Size-selective formation of C78 fullerene from a three-dimensional polyyne precursor

Article abstract of DOI:10.1002/chem.200401040

Multicyclic cagelike cyclophanes 2a and 2b containing cyclobutene rings have been prepared as precursors of three-dimensional polyynes C ₇₈H₁₈ (1a) and C₇₈H₁₂Cl₆ (1b), respectively. Laser irradiation

Full text of DOI:10.1002/chem.200401040

FULL PAPER
Size-Selective Formation of C78 Fullerene from a Three-Dimensional Polyyne
Precursor
Yoshito Tobe,*[a] Rui Umeda,[a] Motohiro Sonoda,[a] and Tomonari Wakabayashi[b]
Abstract: Multicyclic cagelike cyclo-
phanes 2a and 2b containing cyclobu-
tene rings have been prepared as pre-
cursors of three-dimensional polyynes
C78H18 (1a) and C78H12Cl6 (1b), respec-
tively. Laser irradiation of 2a and 2b
induced expulsion of the aromatic frag-
ment, indane, to give the three-dimen-
former anion lost only four hydrogen
atoms to form C78H14À, complete loss
of all hydrogen and chlorine atoms was
observed from the latter anion, to yield
a C78 ion that has a fullerene structure
which was proven by its characteristic
fragmentation pattern.
À
sional polyyne anions C78H18 and
C78H12Cl6À, respectively. Whereas the
À
Keywords: alkynes · cyclophanes ·
fullerenes · macrocycles · polycycles
Introduction
et al. have shown that the appropriately designed polycyclic
aromatic hydrocarbon C60H30 converts into the C60 ion se-
+
Since the discovery of fullerene[1] and its preparative-scale
production,[2] the chemistry of fullerenes has been extensive-
ly developed mainly for the most abundant C60 and C70.
However, little has been done for fullerenes larger than C70,
even though the larger fullerenes are expected to display in-
teresting properties that are not conceivable with C60 and
C70.[3] This is mainly because the larger fullerenes are
formed in low yields as hard-to-isolate mixtures of by-prod-
ucts of the production of C60 and C70 and because there
exists an increasing number of possible isomers with an in-
creasing number of the carbon atoms.[4] The only exception
to this trend is the endohedral fullerenes,[5] most of which
have a carbon cage larger than C82, but there are still a very
limited number of them available. Thus, the size-selective
and geometry-selective formations of large fullerenes are
two important problems hitherto unsolved in fullerene sci-
ence. The chemical synthesis of fullerenes from structurally
well-defined precursors is promising in these respects com-
pared to the nonselective methods such as vaporization of
graphite and combustion of hydrocarbons. Indeed, Scott
lectively upon laser irradiation in the gas phase by C C
À
bond formation accompanying the dehydrogenation.[6]
Moreover, they reported the first size-selective preparation
of C60 by the high-temperature pyrolysis of the related chlor-
ohydrocarbon C60H27Cl3.[7] Cyclic polyynes can be regarded
as viable precursors of C60 fullerene, since Roskamp and Jar-
rold proposed a mechanism for C60 fullerene formation to
explain ring formation between spiraling polyyne chains.[8] It
should be pointed out that they also predicted that similar
polyyne cyclization of appropriate precursors would form
larger fullerenes as well through similar pathways. Experi-
mentally, Diederich and co-workers first observed the size-
+
selective formation of the fullerene C60 by the gas-phase
coalescence of the cyclic polyyne C30+, which was produced
size-selectively from a well-defined precursor.[9] Although a
+
relatively strong peak for the C70 ion was detected by simi-
+
lar coalescence of C18 and C24+, many other carbon clusters
ions were also formed; size-selective formation of larger
fullerenes has not so far been achieved.[10] In connection
with the polyyne route to fullerenes, Rubin and co-workers
and our group have developed independently an approach
to C60 from well-defined, three-dimensional (3D) polyyne
precursors C60H6 and C60Cl6, which were generated from the
corresponding precursors by expulsion of stable fragments
such as carbon monoxide[11] or indane[12] by laser irradiation.
Indeed, the 3D polyynes collapsed into C60 ions (both nega-
tive and positive) accompanied by the loss of hydrogen and/
or chlorine atoms. Even though the preparative-scale syn-
thesis of C60 has not yet been achieved by this route, we
planned to extend it to the size-selective formation of larger
[a] Prof. Dr. Y. Tobe, R. Umeda, Dr. M. Sonoda
Division of Frontier Materials Science
Graduate School of Engineering Science
Osaka University, and CREST (JST)
1–3 Machikaneyama, Toyonaka 560–8531 (Japan)
Fax : (+81)6-6850-6229
E-mail: tobe@chem.es.osaka-u.ac.jp
[b] Prof. Dr. T. Wakabayashi
Department of Chemistry, School of Science and Engineering
Kinki University, Higashi-Osaka 557–8502 (Japan)
Chem. Eur. J. 2005, 11, 1603 – 1609
DOI: 10.1002/chem.200401040
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1603

Results and Discussion
The synthesis of precursors 2a and 2b is shown in
Scheme 2.[15] Selective removal of the trimethylsilyl group
from the diethynylpropellane derivative 3a[14] gave 3b. Sub-
sequent Pd-catalyzed cross-coupling of 3b with diethyl-(4-
iodophenyl)triazene[16] gave compound 4a, and deprotection
of the triisopropylsilyl group gave 4b. Pd-catalyzed cross-
coupling of three equivalents of 4b with the tris(bromoethy-
nyl)benzene derivative 5a afforded the triad 6a, and the
Scheme 1. A speculation for the spiraling polyyne cyclization for C78H18
and C78H12Cl6 to form C78.
fullerenes. Namely, on the basis
of a speculative mechanism for
polyyne cyclization,[12b,13] we en-
visioned that insertion of ben-
zene (shaded in Scheme 1) or
phenylene units into the poly-
yne chain of the C60 precursor
would lead to the formation of
higher fullerenes of C60
+
18n
(n=1,
2 …), as shown in
Scheme 1 for D3-C78 as an ex-
ample. Accordingly, we de-
signed 3D polyynes 1a and 1b
as precursors of C78 and cage-
like cyclophanes 2a and 2b
having [4.3.2]propellane units
as precursors of the polyynes
1a and 1b, respectively.[14] We
disclose here the synthesis of
cyclophanes 2a and 2b, the
generation of polyynes 1a and
1b from them, and the size-se-
lective formation of C78 fuller-
ene from the chlorine-contain-
ing polyyne 1b.
Scheme 2. Synthesis of precursors 2a and 2b of three-dimensional polyynes 1a and 1b.
1604
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1603 – 1609

C78 Fullerene
FULL PAPER
subsequent transformation of the triazene to iodo group
gave 7a. The final cross-coupling reaction of 7a with an in
situ generated terminal alkyne derived from 8a[12] by remov-
al of the triisopropylsilyl group yielded the multicyclic cage-
like compound 2a. Similarly, coupling of 4b with the tri-
chloro derivative 5b to form 6b followed by iodination gave
7b, which was subjected to cross-coupling with an in situ
prepared terminal alkyne derived from 8b[12b] to furnish the
hexachloro derivative 2b.
Laser-desorption (LD) time-of-flight (TOF) mass spectra
of 2a and 2b were recorded by irradiation of the solid sam-
ples with a nitrogen laser (337 nm) followed by detection of
the negative ions in the reflectron mode (Figure 1). Laser
ionization/desorption of 2a leads principally to the forma-
tion of the negative ion of 1a (C78H18À) by extrusion of six
indane fragments from the molecular ion, which is not de-
tected (Figure 1a). The peaks due to the anions
[C78H18(indane)]À and [C78H18(indane)2]À, in which one or
two cyclobutene rings are still intact, are also observed. As
shown in the inset of Figure 1a, the loss of four hydrogen
À
atoms down to C78H14 (m/z 960) was observed. The isotope
distribution of the peaks at m/z 960–962 strongly suggests
À
the absence of intermediate anions such as C78H15 and
C78H16À, indicating that the loss of four hydrogen atoms
takes place simultaneously. However, in contrast to the
À
anion C60H6 which loses six hydrogen atoms to form
À
[11,12]
À
À
C60
,
the anions C78H18 and C78H14 are resistant to fur-
ther elimination of hydrogen, just as observed for
À
[17]
C60H18
,
indicating that the hydrogen losses accompanied
by the polyyne cyclization are not kinetically favorable. In
addition, further fragmentation due to the loss of C2, the
collapse process characteristic to fullerene cations[18] and
À
anions,[19] from C78H14 was not observed. These results
strongly suggest that the C78H14 anion does not possess a
fullerene-like structure. On the other hand, the halogenated
derivatives are known to serve better than the correspond-
ing hydrocarbon as precursors of fullerenes,[7,12b,20] presuma-
À
bly because the facile cleavage of C Cl bonds would pro-
vide favorable funnels for the intramolecular ring closure.
Indeed, expulsion of most of the hydrogen and chlorine
atoms takes place simultaneously from the chlorinated
À
anion C78H12Cl6 generated by laser ionization/desorption of
Figure 1. a) Negative-ion mode laser desorption TOF mass spectrum of
À
2a [C78H18(indane)6]À. Inset: Expansion of the spectrum for C78H18 and
precursor 2b (Figure 1b and c). As shown in the inset of Fig-
ure 1b, the distribution of the peaks at m/z 936–944 indi-
cates the presence of C78 together with C78H2 and C78H4
in which two or four hydrogen atoms are still intact. Fig-
ure 2a shows the observed peak distribution which is nicely
C78H14À. b) Negative-ion mode laser desorption TOF mass spectrum of
2b [C78H12Cl6(indane)6]À at relatively high laser fluence. Inset: Expansion
À
À
À
À
À
of the spectrum for C78 and C78Hn (n=2, 4, 6). c) Negative-ion mode
laser desorption TOF mass spectrum of 2b [C78H12Cl6(indane)6]À at rela-
tively low laser fluence.
À
reproduced by simulation assuming the distribution of C78
,
C78H2À, C78H4À, and C78H6 in a ratio of 10:10:4:1 (Fig-
ure 2b). The relative ratio of these ions did not change upon
increasing the laser fluence. It is rather surprising that the
À
À
À
from C78 down apparently to C70 were observed at rela-
tively high laser fluence, which are characteristic to the full-
erene ions.[18,19] Carbon and hydrocarbon ions larger than
À
À
hydrogen-containing anions such as C78H2 and C78H4 are
resistant to dehydrogenation in spite of their high degree of
unsaturation. It is not clear, however, whether these species
possess fullerene structures. On the other hand, the C78 ion
is more likely to have a fullerene structure; as shown in Fig-
ure 1b, small peaks due to the fragmentation by loss of C2
À
À
C78 up to C86 formed by C2 addition are also observed,
similar to the situation for the C60 anion formed from its
polyyne precursors.[11,12] These fragmentation peaks were
not observed in the mass spectrum at low laser fluence (Fig-
ure 1c). For the lower mass region, the calculated isotope
À
Chem. Eur. J. 2005, 11, 1603 – 1609
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1605

Y. Tobe et al.
À
À
Figure 2. a) Observed peak distribution for C78 and C78Hn (n=2, 4, 6).
b) Calculated peak distribution for C78À, C78H2À, C78H4À, and C78H6 in a
À
ratio of 10:10:4:1.
À
Figure 4. a) Observed spectrum for the higher mass region (C80À, C82
,
C84À, and C86À). b) Calculated peak distributions assuming C80 + C80H +
C80H2 (10:1:3), C82 + C82H (10:3), C84 + C84H + C84H2 (10:3:3), and C86
+ C86H + C86H2 (10:3:3).
À
sults strongly suggest that the C78 ion generated from 1b
possesses a fullerene structure.
Further spectroscopic characterization by, for example, ul-
traviolet photoelectron spectroscopy (UPS) should confirm
this conjecture. For C78 fullerene, there are five possible
geometrical isomers, three of which have been isolated and
identified by the three different research groups.[21] The dif-
ference between the distributions of the isomers reported
from the different groups was interpreted in terms of the
temperature and pressure of a buffer gas, suggesting the
presence of kinetically controlled pathways for the forma-
tion of the isomers rather than their equilibration.[22] Identi-
À
Figure 3. a) Observed spectrum for the lower mass region (C70À, C72
C74À, and C76À). b) Calculated isotope distribution for C70 to C76.
,
À
fication of the geometry of the C78 ion thus formed to clari-
fy the geometrical selectivity is the subject of our current in-
vestigation.
The importance of the well-defined 3D structure for the
efficient cyclization of the polyyne chains is exemplified by
the failure to form C78 from the highly unsaturated acyclic
polyyne 9 (C78H18Cl12) having 78 carbon atoms under similar
conditions. Compound 10,[15] a precursor of polyyne 9, was
prepared by cross-coupling of 7b with (2,4,6-trichlorophen-
yl)butadiyne (Scheme 3). As shown in Figure 5, the nega-
tive-mode laser desorption TOF mass spectrum of 10 exhib-
its strong peaks centered at m/z 1380 corresponding to poly-
distributions for C76 to C70 (Figure 3b) fit the observed spec-
trum (Figure 3a), indicating that the C2 loss occurs exclusive-
À
ly from the C78 ion. On the other hand, the peak distribu-
tions for the higher mass region (Figure 4b) calculated by
assuming C80 + C80H + C80H2 (10:1:3), C82 + C82H (10:3),
C84 + C84H + C84H2 (10:3:3), and C86 + C86H + C86H2
(10:3:3) fit the observed spectrum (Figure 4a), indicating
À
that the C2 uptake occurs to both C78 and C78H2À. These re-
1606
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1603 – 1609

C78 Fullerene
FULL PAPER
In summary, we have synthesized the multicyclic com-
pounds 2a and 2b containing cyclobutene rings as precur-
sors of three-dimensional polyynes, C78H18 and C78H12Cl6.
Laser irradiation of 2a and 2b induced expulsion of the aro-
À
matic fragment, indane, giving the polyyne anions C78H18
and C78H12Cl6À, respectively, and subsequent cyclization of
the polyyne chain of the latter anion gave rise to the size-se-
lective formation of C78À. Since it should be possible to con-
struct well-defined polyyne precursors larger than 2a and
2b, this method would be applicable to the size-selective for-
mation of even larger, tubular fullerenes.
Experimental Section
General: 1H and 13C NMR spectra were recorded on a Varian Mercury
300, a JEOL JNM AL-400, or a Varian Unity Inova 500 spectrometer in
CDCl3 and with Me4Si or residual solvent as an internal standard at
308C. The multiplicity in the 13C NMR spectra was determined by DEPT
spectra and was indicated by (s), (d), (t), and (q) for quaternary, tertially,
secondary, and primary carbon atoms, respectively. IR spectra were re-
corded as KBr disks with a JASCO FTIR-410 spectrometer. Mass spec-
tral analyses were performed on a JEOL JMS-700 spectrometer for EI
and FAB ionization. The LD-TOF mass measurements were recorded on
a
Shimadzu/Kratos AXIMA-CFR spectrometer. Melting points were
measured with a hot-stage apparatus and are uncorrected. Column chro-
matography and TLC were performed with Merck silica gel 60 (70–
230 mesh ASTM) and Merck silica gel 60 F254, respectively. Preparative
HPLC was undertaken with
a JAI LC-908 chromatograph using
600 mmꢁ20 mm JAIGEL-1H and 2H GPC columns with CHCl3 as the
eluent. All reagents were obtained from commercial suppliers and used
as received. Solvents were dried (drying agent in parentheses) and distil-
led prior to use: THF (LiAlH4 followed by sodium benzophenone ketyl),
benzene (CaH2), Et3N (KOH), Et2O (CaH2).
Scheme 3. Synthesis of precursors 10 of acyclic polyyne 9.
Triazene 4a: A solution of 3a[14] (3.00 g, 7.13 mmol) in THF (50 mL) was
added to a solution of K2CO3 (2.02 g, 14.6 mmol) in MeOH (80 mL) with
stirring at room temperature. After being stirred for 1 h, the reaction
mixture was filtered and diluted with Et2O. The solution was washed
with brine, dried over anhydrous MgSO4, and concentrated in vacuo to
give the deprotected product which was used without purification. The
crude sample was dissolved in Et3N (ca. 20 mL), which had been de-
gassed thoroughly by bubbling argon for 1 h, and the solution was added
dropwise under an argon atmosphere to a solution of [PdCl2(PPh3)2]
(443 mg, 383 moml), CuI (128 mg, 674 moml), and N,N-diethyl-N’-(4-iodo-
phenyl)triazene[16] (2.18 g, 7.19 mmol) in degassed Et3N (25 mL) over 1 h.
After being stirred for 2 h, the reaction mixture was filtered and the fil-
trate was concentrated in vacuo. The residue was passed through a short
plug of silica gel (n-hexane/EtOAc 9:1). Purification by preparative
HPLC afforded triazene 4a (3.19 g, yield 85% for two steps) as an
1
orange oil: H NMR (300 MHz, CDCl3): d=7.41–7.33 (m, 4H), 5.89–5.85
(m, 4H), 3.77 (q, J=7.0 Hz, 4H), 2.60–1.94 (m, 2H), 1.64–1.43 (m, 2H),
1.31–1.17 (m, 2H), 1.27 (t, J=7.0 Hz, 6H), 1.10 ppm (brs, 21H);
13C NMR (75 MHz, CDCl3): d=151.24 (s), 132.70 (s), 132.62 (d), 132.49
(s), 128.96 (d), 128.90 (d), 121.73 (d), 121.55 (d), 120.33 (d), 119.02 (s),
99.11 (s), 97.81 (s), 95.47 (s), 82.09 (s), 56.02 (s), 55.74 (s), 33.03 (t), 32.87
(t), 18.82 (t), 18.66 (q), 11.25 ppm (d) (the signals of diethyltriazenyl
carbon atoms were not observed because of extensive broadening.); IR
(KBr): n˜ =3026, 2941, 2864, 2189, 2126, 1573, 1463, 1432, 1396, 1329,
1238, 1108, 840, 761 cmÀ1; MS (EI): m/z 523 [M+]; HRMS (EI) calcd for
C34H45N3Si 523.3383, found 523.3403.
Figure 5. Negative-ion mode laser desorption TOF mass spectra of 10
[C78H18Cl12(indane)3], exhibiting peaks due to 9 (C78H18Cl12À) at m/z cen-
tered at 1380.
yne 9 formed by extrusion of three indane units from 10.
However, the loss of only one chlorine atom, forming
À
C78H18Cl11 (m/z centered at 1344), was observed from 9.
Trispropellane 6a:
A solution of tetra-n-butylammonium fluoride
Further losses of hydrogen and chlorine atoms are not ob-
served at all even under high laser fluence, apparently indi-
cating the absence of kinetically favored pathways for the
polyyne cyclization from ill-defined polyynes such as 9.
(TBAF) (1m in THF, 3.9 mL, 3.9 mmol) was added dropwise to a solution
of propellane 4a (1.01 g, 1.98 mmol) in THF (50 mL). After being stirred
for 1 h, the mixture was diluted with water, and then THF was removed
by evaporation. The mixture was extracted with Et2O, and the extract
Chem. Eur. J. 2005, 11, 1603 – 1609
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1607

Y. Tobe et al.
was washed with brine and dried over anhydrous MgSO4. The solvent
was removed by evaporation to give 4b which was used in the next step
without purification. The crude samples of 4b and 5a[14] (194 mg,
502 mmol) were dissolved in benzene (21 mL), and the solution was
added dropwise to a solution of [Pd2(dba)3]·CHCl3 (52.1 mg, 50.3 m mol),
CuI (11.3 mg, 59.3 mmol), and 1,2,2,6,6-pentamethylpiperidine (PMP)
(360 mL, 309 mg, 1.99 mmol) in benzene (8.0 mL) over 15 min at room
temperature. After being stirred for 2 h, the reaction mixture was con-
centrated and passed through a short plug of silica gel (Et2O). The eluate
was washed with 0.5n HCl. The organic layer was washed with saturated
NaHCO3 solution and brine, and dried over anhydrous MgSO4. After re-
moval of the solvent under reduced pressure, purification by preparative
HPLC afforded 6a (562 mg, yield 90% for two steps) as a yellow solid:
m.p. 141–1438C; 1H NMR (300 MHz, CDCl3): d=7.55 (s, 3H), 7.45 and
7.37 (AA’BB’, JA,B =8.4 Hz, 12H), 5.92–5.88 (m, 12H), 3.76 (q, J=
7.2 Hz, 12H), 2.04–2.00 (m, 6H), 1.76–1.50 (m, 6H), 1.35–1.23 (m, 6H),
1.26 ppm (t, J=7.2 Hz, 18H); 13C NMR (75 MHz, CDCl3) d=151.59 (s),
137.37 (s), 136.16 (d), 132.86 (d), 129.16 (s), 128.48 (d), 123.08 (s), 122.03
(d), 121.89 (d), 120.38 (d), 118.35 (s), 97.80 (s), 82.03 (s), 81.17 (s), 78.33
(s), 76.02 (s), 75.48 (s), 56.57 (s), 56.42 (s), 33.09 (t), 33.06 (t), 18.80 ppm
(t) (the signals of diethyltriazenyl carbons were not observed because of
extensive broadening.); IR (KBr): n˜ =3027, 2925, 2850, 2172, 2126, 1578,
1395, 1330, 1237, 1201, 1159, 1079, 842, 737, 734, 680 cmÀ1; MS (FAB): m/
z 1246 [M+].
tion to give a crude sample of 4b. This product 4b and 5b[14] (93.8 mg,
191 mmol) were dissolved in benzene (10 mL), which had been degassed
thoroughly by bubbling argon for 1 h, and the solution was added drop-
wise to a solution of [Pd2(dba)3]·CHCl3 (20.5 mg, 19.8 mmol), CuI (3.7 mg,
19 mmol), PMP (150 mL, 129 mg, 839 mmol) in degassed benzene (10 mL)
over 15 min at room temperature. After being stirred for 2 h, the reaction
mixture was concentrated and passed through a short plug of silica gel
(Et2O). The eluate was washed with 0.5n HCl (100 mL). The organic
layer was washed with saturated NaHCO3 solution and brine, and dried
over anhydrous MgSO4. After removal of the solvent under reduced
pressure, purification by preparative HPLC afforded 6b (151 mg, yield
58% for two steps) as
a
yellow solid: m.p. 142–1448C; 1H NMR
(300 MHz, CDCl3): d=7.47 and 7.38 (AA’BB’, JA,B =8.5 Hz, 12H), 5.92
(s, 12H), 3.77 (q, J=7.2 Hz, 12H), 2.12–2.02 (m, 6H), 1.71–1.47 (m, 6H),
1.36–1.20 (m, 6H), 1.27 ppm (t, J=7.2 Hz, 18H); 13C NMR (75 MHz,
CDCl3) d=151.66 (s), 141.00 (s), 138.28 (s), 132.90 (d), 128.64 (s), 128.45
(d), 128.35 (d), 128.30 (d), 122.33 (s), 122.11 (d), 121.93 (d), 120.40 (d),
118.23 (s), 98.38 (s), 85.40 (s), 82.10 (s), 78.69 (s), 77.89 (s), 75.67 (s),
56.77 (s), 56.51 (s), 33.16 (t), 33.08 (t), 18.82 ppm (t) (the signals of dieth-
yltriazenyl carbons were not observed because of extensive broadening.);
IR (KBr): n˜ =3026, 2933, 2872, 2850, 2178, 1577, 1423, 1395, 1328, 1237,
1201, 1158, 1096, 841, 774, 723, 674 cmÀ1; MS (FAB): m/z 1346–1352 (the
most abundant peak at 1349, [M+]).
Iodopropellane 7b: A solution of triazene 6b (447 mg, 331 mmol) and I2
(59 mg, 0.23 mmol) in MeI (50 mL) was charged to a pressure bottle. The
reaction mixture was degassed thoroughly by bubbling argon for 15 min,
and then the vessel was sealed with a screw cap and heated at 1008C for
63 h. After the reaction mixture was cooled to room temperature, the sol-
vent was removed under reduced pressure. The residue was diluted with
Et2O, and the solution was washed with aqueous sodium thiosulfate
(10%) and brine, and dried over anhydrous MgSO4. After the solvent
was concentrated in vacuo, the residue was dissolved in CHCl3 and n-
hexane was added to this solution to give a yellow precipitate, which was
collected by filtration and washed with n-hexane to give 7b (369 mg) as a
yellow solid. The filtrate was concentrated in vacuo, the residue was puri-
fied by chromatography on silica gel (n-hexane/CHCl3 8:2) to afford an
additional sample of 7b (40 mg, combined yield 86%): decomposed grad-
ually from about 1708C; 1H NMR (300 MHz, CDCl3): d=7.68 and 7.21
(AA’BB’, JA,B =8.2 Hz, 12H), 5.98–5.84 (m, 12H), 2.10–1.98 (m, 6H),
1.72–1.44 (m, 6H), 1.36–1.24 ppm (m, 6H); 13C NMR (75 MHz, CDCl3)
d=141.15 (s), 137.58 (d), 137.37 (s), 133.44 (d), 130.12 (s), 128.22 (d),
128.09 (d), 122.30 (d), 122.12 (d), 121.78 (s), 96.06 (s), 95.28 (s), 85.26 (s),
83.22 (s), 78.32 (s), 78.25 (s), 77.20 (s), 75.81 (s), 56.83 (s), 56.79 (s), 33.10
(t), 33.04 (t), 18.82 ppm (t); IR (KBr): n˜ =3026, 2945, 2847, 2850, 2186,
1569, 1481, 1389, 1317, 1057, 1006, 846, 819, 774, 740, 674 cmÀ1; MS
(FAB): m/z 1428–1433 (the most abundant peak at 1430, [M+]).
Iodopropellane 7a: A solution of triazene 6a (491 mg, 394 mmol) and I2
(72.1 mg, 284 mmol) in MeI (55 mL) was charged to a pressure bottle.
The reaction mixture was degassed thoroughly by bubbling argon for
15 min, and then the vessel was sealed with a screw cap and heated at
1008C for 34 h. After the reaction mixture had cooled to room tempera-
ture, the solvent was removed under reduced pressure. The residue was
diluted with Et2O, and the solution was washed with aqueous sodium thi-
osulfate (10%) and brine, and dried over anhydrous MgSO4. After the
solvent was removed in vacuo, the product was purified by chromatogra-
phy on silica gel (n-hexane/CHCl3 9:1) to give 7a (379 mg, 73%) as a
yellow solid: m.p. 124–1268C; 1H NMR (300 MHz, CDCl3): d=7.67 and
7.21 (AA’BB’, JA,B =8.3 Hz, 12H), 7.55 (s, 3H), 5.95–5.85 (m, 12H), 2.10–
1.98 (m, 6H), 1.70–1.43 (m, 6H), 1.36–1.22 ppm (m, 6H); 13C NMR
(75 MHz, CDCl3) d=137.55 (s), 137.37 (d), 136.48 (s), 136.23 (d), 133.41
(d), 130.64 (s), 128.24 (d), 128.20 (d), 123.03 (s), 122.21 (d), 122.01 (d),
121.88 (d), 95.55 (s), 95.14 (s), 83.22 (s), 81.36 (s), 78.73 (s), 75.92 (s),
75.15 (s), 56.73 (s), 56.56 (s), 33.02 (t), 18.81 ppm (t); IR (KBr): n˜ =3024,
2943, 2847, 2184, 2123, 1575, 1481, 1389, 1056, 1006, 905, 877, 861, 818,
742, 680 cmÀ1; MS (FAB): m/z 1327 [M+ + H].
Cage compound 2a: A solution of TBAF (1m, 64 mL, 64 mmol) in THF
(17 mL), which had been degassed thoroughly by bubbling argon for 1 h,
was dropwise by using a syringe pump (0.17 mLminÀ1) to a solution of
[Pd(PPh3)4] (38.4 mg, 33.2 mmol), CuI (4.7 mg, 25 mmol), 8a[14] (34.7 mg,
29.2 mmol), and 7a (38.2 mg, 28.8 mmol) in degassed Et3N (20 mL) under
an argon atmosphere at room temperature. After being stirred for 1 h,
HCl (ca. 200 mL, 0.5n) was slowly poured at 08C, and the reaction mix-
ture was extracted with Et2O. The extract was washed with saturated
NaHCO3 solution and brine, and dried over anhydrous MgSO4. After re-
moval of the solvent, the residue was subjected to chromatography on
silica gel (n-hexane/AcOEt 7:3). The product was further purified by
preparative HPLC to afford 2a (2.7 mg, 11%) as a yellow solid: decom-
Cage compound 2b: A solution of TBAF (1m, 136 mL, 136 moml) in THF
(17 mL), which had been degassed thoroughly by bubbling argon 1 h, was
dropwise by a syringe pump (0.17 mLminÀ1) to a solution of [Pd(PPh3)4]
(36.0 mg, 31.2 mmol), CuI (6.1 mg, 32 mmol), 8b[14] (42.4 mg, 32.8 moml),
and 7b (46.8 mg, 32.7 moml) in degassed Et3N (27 mL) under an argon at-
mosphere at room temperature. After the mixture had been stirred for
1 h, HCl (ca. 200 mL, 0.5n) was slowly added, and the reaction mixture
was extracted with CHCl3. The extract was washed with saturated
NaHCO3 solution and brine, and dried over anhydrous MgSO4. After re-
moval of the solvent, the residue was subjected to chromatography on
silica gel (n-hexane/CHCl3 1:1). The product was further purified by
preparative HPLC to afford 2b (4.0 mg, 7%) as a yellow solid: decom-
posed gradually from 1508C; 1H NMR (300 MHz, CDCl3): d=7.43 (s,
12H), 5.95–5.85 (m, 24H), 2.10–1.98 (m, 12H), 1.72–1.44 (m, 12H), 1.38–
1.24 ppm (m, 12H); 13C NMR (125 MHz, CDCl3) d=141.40 (s), 137.53
(s), 132.03 (d), 130.34 (s), 128.19 (d), 128.09 (d), 122.78 (s), 122.26 (s),
122.13 (d), 96.54 (s), 85.13 (s), 84.22 (s), 78.25 (s), 78.15 (s), 75.58 (s),
56.79 (s), 56.66 (s), 33.12 (t), 33.03 (t), 18.81 ppm (t); IR (KBr): n˜ =3028,
2948, 2850, 2191, 1391, 1075, 836, 774, 695 cmÀ1; LD TOF-MS (Figure 1b
and c).
1
posed gradually from about 1408C; H NMR (400 MHz, CDCl3): d=7.54
(s, 6H), 7.42 (s, 12H), 5.94–5.85 (m, 24H), 2.10–1.96 (m, 12H), 1.72–1.44
(m, 12H), 1.38–1.20 ppm (m, 12H); 13C NMR (75 MHz, CDCl3) d=
136.66 (s), 136.15 (s), 131.76 (d), 130.74 (s), 128.05 (d), 128.02 (d), 122.86
(s), 122.65 (s), 122.03 (d), 121.90 (d), 96.05 (s), 84.17 (s), 81.22 (s), 78.67
(s), 75.77 (s), 75.02 (s), 56.77 (s), 56.56 (s), 33.20 (t), 19.02 ppm (t); IR
(KBr): n˜ =3026, 2932, 2850, 2187, 1577, 1438, 1375, 1076, 876, 836, 756,
734, 691 cmÀ1; LD TOF-MS (Figure 1a).
Trispropellane 6b: A solution of TBAF (1m in THF, 1.5 mL, 1.5 mmol)
was added dropwise to a solution of propellane 4a (320 mg, 0.611 mmol)
in THF (15 mL). After being stirred for 1 h, the mixture was diluted with
water and extracted with Et2O. The extract was washed with brine, and
dried over anhydrous MgSO4, and the solvent was removed by evapora-
1-(2,4,6-Trichlorophenyl)-4-(trimethysilyl)butadiyne:[23]
A
solution of
MeLi–LiBr in Et2O (1.5m, 0.75 mL, 1.13 mol) was added to a solution of
1608
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1603 – 1609

C78 Fullerene
FULL PAPER
1,4-bis(trimethylsilyl)butadiyne (203 mg, 1.04 mmol) in Et2O (6.0 mL)
with stirring at 08C, and then mixture was slowly warmed up room tem-
perature. After being stirred for 24 h, the reaction mixture was slowly
poured into a saturated aqueous NH4Cl solution cooled to 08C (ca.
50 mL). The aqueous phase was extracted with Et2O, and the organic
layer was dried over anhydrous Na2SO4. The solvent was removed under
reduced pressure below room temperature, and the residue was dissolved
in Et3N (ca. 6 mL), which had been degassed thoroughly by bubbling
argon for 1 h. The above solution was added under an argon atmosphere
into a solution of [Pd2(dba)3]·CHCl3 (63.3 mg, 61.2 mmol), CuI (20.2 mg,
106 mmol), PPh3 (84.2 mg, 321 mmol), and 1,3,5-trichroro-2-iodobenzene
(115 mg, 374 mmol) in degassed Et3N (1.0 mL) at 508C. After being stir-
red for 4 h, the reaction mixture was filtered and the filtrate was concen-
trated in vacuo. The residue was passed through a short plug of silica gel
(n-hexane). Purification by preparative HPLC afforded the butadiyne de-
rivative (63.4 mg, yield 56% for two steps) as a colorless solid; m.p. 64–
658C; 1H NMR (300 MHz, CDCl3): d=7.33 (s, 2H), 0.26 ppm (s, 9H);
13C NMR (75 MHz, CDCl3): d=138.86 (s), 134.92 (s), 127.67 (d), 120.61
(s), 94.90 (s), 86.93 (s), 84.63 (s), 69.06 (s), À0.23 ppm (q); IR (KBr): n˜ =
3113, 3076, 2957, 2925, 2853, 2101, 1561, 1536, 1449, 1385, 1371, 1251,
1090, 1019, 974, 849, 761, 655 cmÀ1; MS (EI): m/z 304, 302, 300 (rel. inten-
sity ca. 1:2:2, [M+]); HRMS (EI) calcd for C13H1135Cl37Cl2Si,
C13H1135Cl237Cl1Si, and C13H1135Cl3Si 303.9640, 301.9667, and 299.9696,
found 303.9643, 301.9658, and 299.9679, respectively.
[5] H. Shinohara in Fullerenes: Chemistry, Physics, and Technology,
(Eds.: K. M. Kadish, R. S. Ruoff), Wiley-Interscience, New York,
2000, pp. 357–393.
[6] M. M. Boorum, Y. V. Vasil’ev, T. Drewello, L. T. Scott, Science 2001,
294, 828–831.
[7] L. T. Scott, M. M. Boorum, B. J. McMahon, S. Hagen, J. Mack, J.
Blank, H. Wegner, A. de Meijere, Science 2002, 295, 1500–1503.
[8] J. M. Hunter, J. L. Fye, E. J. Roskamp, M. F. Jarrold, J. Phys. Chem.
1994, 98, 1810–1818.
[9] S. W. McElvany, M. M. Ross, N. S. Goroff, F. Diederich, Science
1993, 259, 1594–1596.
[10] C62À, a non-classical fullerene anion larger than C60À, has been gen-
erated in a size- and geometry-selective manner by formal addition
of a C2 unit to the already existing C60 fullerene core by Rubin et al.
from its precursor C60H2(CO)2 under laser desorption mass spectro-
metric conditions. Though selective, the range of larger fullerenes
produced by this method should be limited compared to those de-
rived from non-fullerene precursors: W. Quian, M. D. Bartberger,
S. J. Pastor, K. N. Houk, C. L. Wilkins, Y. Rubin, J. Am. Chem. Soc.
2000, 122, 8333–8334.
[11] Y. Rubin, T. C. Parker, S. J. Pastor, S. Jalisatgi, C. Boulle, C. L. Wil-
kins, Angew. Chem. 1998, 110, 1353–1356; Angew. Chem. Int. Ed.
1998, 37, 1226–1229.
[12] a) Y. Tobe, N. Nakagawa, K. Naemura, T. Wakabayashi, T. Shida, Y.
Achiba, J. Am. Chem. Soc. 1998, 120, 4544–4545; b) Y. Tobe N. Na-
kagawa, J. Kishi, M. Sonoda, K. Naemura, T. Wakabayashi, T.
Shida, Y. Achiba, Tetrahedron 2001, 57, 3629–3636.
Acyclic polyyne precursor 10:
A solution of the above butadiyne
(55.3 mg, 183 mmol) in THF (3.0 mL) was added to a solution of K2CO3
(46.7 mg, 338 mmol) in MeOH (3.0 mL) with stirring at room tempera-
ture. After being stirred for 1 h, the reaction mixture was filtered and di-
luted with Et2O. The solution was washed with brine, dried over anhy-
drous MgSO4, and concentrated in vacuo to give a crude sample of the
deprotection product. The crude product was dissolved in Et3N (ca.
8 mL), which had been degassed thoroughly by bubbling argon for 1 h,
and the solution was added under an argon atmosphere to a solution of
[Pd(PPh3)4] (14.5 mg, 12.5 mmol), CuI (3.2 mg, 17 mmol), and 7b (53.5 mg,
37.4 mmol) in degassed Et3N (3.0 mL) at room temperature. After being
stirred for 2 h, the reaction mixture was filtered and the filtrate was con-
centrated in vacuo. The residue was passed through a short plug of silica
gel (n-hexane/CHCl3 1:1). Purification by preparative HPLC afforded 10
(59.9 mg, yield 92% for two steps) as a brown solid: decomposed gradu-
ally from about 1408C; 1H NMR (300 MHz, CDCl3): d=7.50 and 7.46
(AA’BB’, JAB =7.3 Hz, 12H), 7.34 (s, 6H), 5.95–5.84 (m, 12H), 2.10–1.99
(m, 6H), 1.72–1.43 (m, 6H), 1.37–1.22 ppm (m, 6H); 13C NMR (75 MHz,
CDCl3) d=140.96 (s), 138.66 (s), 136.95 (s), 134.96 (s), 132.30 (d), 131.87
(d), 130.31 (s), 127.99 (d), 127.87 (d), 127.70 (d), 123.41 (s), 122.13 (d),
122.16 (d), 122.00 (d), 121.55 (s), 96.20 (s), 85.20 (s), 84.62 (s), 84.55 (s),
84.26 (s), 78.47 (s), 78.27 (s), 77.20 (s), 75.93 (s), 75.58 (s), 75.24 (s), 56.98
(s), 56.91 (s), 33.29 (t), 33.22 (t), 19.04 ppm (t); IR (KBr): n˜ =3026, 2943,
2848, 2207, 1601, 1572, 1533, 1448, 1387, 1371, 1318, 1185, 1139, 1081,
1016, 857, 836, 818, 774, 759, 702, 674 cmÀ1; LD TOF-MS (Figure 5).
[13] U. H. F. Bunz, Y. Rubin, Y. Tobe, Chem. Soc. Rev. 1999, 28, 107–
119.
[14] For the formation of reactive polyynes by [2+2] cycloreversion of
[4.3.2]propellane derivatives, see: Y. Tobe, T. Fujii, H. Matsumoto,
K. Tsumuraya, D. Noguchi, N. Nakagawa, M. Sonoda, K. Naemura,
Y. Achiba, T. Wakabayashi, J. Am. Chem. Soc. 2000, 122, 1762–
1775.
[15] Compounds 2a,b, 6a,b–8a,b, and 10 are mixtures of diastereomers.
The structure of only one of each of them is drawn for clarity.
[16] B. T. Holmes, W. T. Pennington, T. W. Hanks, Synth. Commun. 2003,
33, 2447–2461.
[17] Y. Rubin, T. C. Parker, S. I. Kahn, C. L. Holiman, S. W. McElvany, J.
Am. Chem. Soc. 1996, 118, 5308–5309.
[18] a) S. C. OꢂBrien, J. R. Heath, R. F. Curl, R. E. Smalley, J. Chem.
Phys. 1988, 88, 220–230; b) P. P. Radi, T. L. Bunn, P. R. Kemper,
M. E. Molchan, M. T. Bowers, J. Chem. Phys. 1988, 88, 2809–2814;
c) T. Wakabayashi, H. Shiromaru, S. Suzuki, K. Kikuchi, Y. Achiba,
Surf. Rev. Lett. 1996, 3, 793–798.
[19] N. G. Gotts, G. von Helden, M. T. Bowers, Int. J. Mass Spectrom. Ion
Processes 1995, 149/150, 217–229.
[20] Y. Tobe, R. Furukawa, M. Sonoda, T. Wakabayashi, Angew. Chem.
2001, 113, 4196–4198; Angew. Chem. Int. Ed. 2001, 40, 4072–4074.
[21] a) F. Diederich, R. L. Whetten, C. Thilgen, R. Ettl, I. Chao, M. M.
Alvarez, Science 1991, 254, 1768–1770; b) K. Kikuchi, N. Nakahara,
T. Wakabayashi, S. Suzuki, H. Shiromaru, Y. Miyake, K. Saito, I.
Ikemoto, M. Kainosho, Y. Achiba, Nature 1992, 357, 142–145; c) R.
Taylor, G. J. Langley, T. J. S. Dennis, H. W. Kroto, D. R. M. Walton,
J. Chem. Soc. Chem. Commun. 1992, 1043–1046.
Acknowledgement
Support to R. U. from the 21st Century COE Program “Integrated Eco-
Chemistry” is gratefully acknowledged.
[22] T. Wakabayashi, D. Kasuya, H. Shiromaru, S. Suzuki, K. Kikuchi, Y.
Achiba, Z. Phys. D 1997, 40, 414–417.
[23] (Trimethylsilyl)butadiyne was prepared according to the procedure
reported by Gladysz and co-workers: B. Bartik, R. Dembinski, T.
Bartik, A. M. Arif, J. A. Gladysz, New J. Chem. 1997, 21, 739–750.
[1] H. W. Kroto, J. R. Heath, S. C. OꢂBrien, R. F. Curl, R. E. Smalley,
Nature 1985, 318, 162–163.
[2] W. Krꢃtschmer, L. D. Lamb, K. Fostiropoulos, D. R. Huffman,
Nature 1990, 347, 354–358.
[3] F. Diederich, R. L. Whetten, Acc. Chem. Res. 1992, 25, 119–126.
[4] P. W. Fowler, D. E. Manolopoulos, Atlas of Fullerenes, Oxford Uni-
versity Press, Oxford, 1995.
Received: October 13, 2004
Published online: January 20, 2005
Chem. Eur. J. 2005, 11, 1603 – 1609
www.chemeurj.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1609