Precursors to endohedral metal fullerene complexes: Synthesis and X-ray structure of a flexible acetylenic cyclophane C60H18

Full text of DOI:10.1021/ja9606384

5308
J. Am. Chem. Soc. 1996, 118, 5308-5309
Precursors to Endohedral Metal Fullerene
Complexes: Synthesis and X-ray Structure of a
Flexible Acetylenic Cyclophane C60H18
Yves Rubin,*,† Timothy C. Parker,† Saeed I. Khan,†
Christopher L. Holliman,‡ and Stephen W. McElvany*,‡
Department of Chemistry and Biochemistry
UniVersity of California
Los Angeles, California 90095-1569
NaVal Research Laboratory, Chemistry DiVision
Washington, D.C. 20375
ReceiVed February 27, 1996
Figure 1. ORTEP plot of the crystal structure of 7.
The rational synthesis of stable endohedral metal fullerene
complexes poses a major challenge to current synthetic meth-
odologies. We are particularly interested in C60 endohedral
complexes because C60 itself has delivered the most interesting
physical properties among the empty fullerenes.1 While lan-
thanide endohedral complexes of higher fullerenes (e.g. La@C82)
have been available for some time by resistive evaporation
methods, analogous C60 complexes do not seem to be isolable.2
Endohedral complexes of C60 with noble gases (e.g. He@C60)
have been prepared by insertion into empty fullerenes under
high pressure and temperature.3 These methods are limited by
the choice of guest atom, fullerene size, and very low yields.
We are proposing an unconventional approach to fullerene
synthesis from polyalkynyl precursors such as 1 and 2.4 This
approach is based on the remarkable propensity of cyclo[n]-
carbons to rearrange to fullerenes in the gas phase (e.g. 3
pyrolytic conditions or even in solution.8,9 A metal guest
included in their cavity should be trapped endohedrally when
rearrangement occurs. We are reporting the synthesis, X-ray
structure, and preliminary mass spectroscopic study of cyclo-
phane 7 in our first approach to the rational synthesis of
fullerenes.
The short convergent route to cyclophane 7 was devised by
application of straightforward alkynylation procedures. The tris-
(bromoalkyne) 4b was prepared from 1,3,5-triethynylbenzene
(4a)10 by bromination with NBS.11 The copper acetylide 5 was
obtained by selectively deprotecting 1-(triisopropylsilyl)-6-
(trimethylsilyl)-3-hexen-1,5-diyne12 with K2CO3/MeOH, fol-
lowed by deprotonation (LHMDS, THF) and addition of CuBr.
A three-fold coupling of 5 to 4b in pyridine13 yielded the stable
intermediate 6a which was deprotected (TBAF) to give the
unstable triyne 6b. Cyclization of 6b under Hay conditions in
o-dichlorobenzene (ODCB) afforded the bright yellow cyclo-
phane 7 in 34% yield. Cyclophane 7 decomposes over a period
of weeks in the crystalline state but is stable indefinitely in dilute
solutions (e.g. CH2Cl2) if kept in the dark.
converting to C60).5-7 Even though the mechanism of this
rearrangement is complex, we envision 60-carbon polyacetylenic
cyclophanes such as 1 and 2 analogously forming C60 after being
generated from more stable precursors under flash vacuum
After considerable effort, crystals suitable for X-ray diffrac-
tion were grown by vapor phase diffusion of heptane into a
solution of 7 in ODCB (Figure 1). Analysis of the crystal
structure14 reveals that the two benzene rings of cyclophane 7
are not stacked on top of each other, but rather are offset by
† University of California.
‡ Naval Research Laboratory.
(8) The AM1 calculated heat of formation of the acetylenic precursor 2
is 562 kcal/mol (9.4 kcal/mol per carbon) higher than its isomer C60.
(9) For the remarkable “zipper” rearrangement of cyclic o-ethynylben-
zenes, see: Bradshaw, J. D.; Solooki, D.; Tessier, C. A.; Youngs, W. J. J.
Am. Chem. Soc. 1994, 116, 3177-3179.
(1) Hirsch, A. The Chemistry of the Fullerenes; Thieme Verlag: Stuttgart,
1994.
(2) Edelmann, F. T. Angew. Chem., Int. Ed. Engl. 1995, 34, 981-985
and references therein.
(3) Saunders, M.; Cross, R. J.; Jime´nez-Va´zquez, H. A.; Shimshi, R.;
Khong, A. Science 1996, 271, 1693-1697.
(10) Weber, E.; Hecker, M.; Koepp, E.; Orlia, W.; Czugler, M.; Cso¨regh,
I. J. Chem. Soc., Perkin Trans. 2 1988, 1251-1257.
(4) Most approaches to C60 reported so far rely on the synthesis of bowl-
shaped fragments such as corannulene, see: (a) Abdourazak, A. H.;
Marcinow, Z.; Sygula, A.; Sygula, R.; Rabideau, P. W. J. Am. Chem. Soc.
1995, 117, 6410-6411. (b) Faust, R. Angew. Chem., Int. Ed. Engl. 1995,
34, 1429-1432 and references cited therein.
(11) Hofmeister, H.; Annen, K.; Laurent, H.; Wiechert, R. Angew. Chem.,
Int. Ed. Engl. 1984, 23, 727-729.
(12) Lu, Y. F.; Harwig, C. W.; Fallis, A. G. J. Org. Chem. 1993, 58,
4202-4204.
(13) Miller, J. A.; Zweifel, G. Synthesis 1983, 128-130.
(14) Compound 7 (C60H18‚3o-C6H4Cl2; Mr ) 1179.81) crystallized in
the orthorhombic space group Pbca with cell dimensions of a ) 35.320(5)
Å, b ) 40.488(7) Å, and c ) 8.458(2) Å; V ) 12095(4) Å3, and an
occupation of Z ) 8 in the unit cell. Data were collected at 110 K on a
AFC5R Rigaku diffractometer using graphite-monochromated Cu KR
radiation, to a maximum 2θ ) 120°, giving 10 054 unique reflections; the
structure was solved by direct methods (SHELX86) and refined with full
matrix least squares, yielding R ) 0.062, Rw ) 0.080 for 5383 independent
reflections with I > 3σ(I).
(5) (a) Rubin, Y.; Kahr, M.; Knobler, C. B.; Diederich, F.; Wilkins, C.
L. J. Am. Chem. Soc. 1991, 113, 495-500. (b) McElvany, S. W.; Ross, M.
M.; Goroff, N. S.; Diederich, F. Science 1993, 259, 1594-1596. (c) Goroff,
N. S. Acc. Chem. Res. 1996, 29, 77-83.
(6) von Helden, G.; Gotts, N. G.; Bowers, M. T. Nature 1993, 363, 60-
63.
(7) (a) Clemmer, D. E.; Jarrold, M. F. J. Am. Chem. Soc. 1995, 117,
8841-8850. (b) Hunter, J.; Fye, J.; Jarrold, M. F. Science 1993, 260, 784-
786.
S0002-7863(96)00638-5 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 22, 1996 5309
Table 1. Calculated Barriers (kcal/mol) for the Racemization of 7
AM116
PM316
CVFF17
CFF9117
+3.28
+2.81
+6.13
+3.37
Figure 2. Space-filling model for a partial packing diagram showing
two pairs of enantiomeric units of 7 and four ODCBs: light gray )
right handed, dark gray ) left handed.
∼1.32 Å along the C23-C26 axis (offset distances between
C2-C28 and C6-C24 are 1.254 and 1.386 Å, respectively).
The two pairs of carbons C2-C27 and C6-C25 are essentially
sitting on top of each other as a result. The angle between the
mean planes of the two benzene rings is 5.0°. The distances
between the two planes, as defined by the normal from C2 and
C6 to the opposite plane, are 3.323 and 3.250 Å, essentially
within van der Waals distances. In addition, the butadiynyl
moieties are bent as much as 5° from their ideal linearity, which
imparts C1 symmetry to the whole structure in the crystalline
state. The fact that the two benzene rings are not superimposed
may be due to crystal-packing forces combined with electrostatic/
π-stacking effects. The distortion of 7 in the solid state from
more energetically favorable structures found by minimization
is not surprising due to the relative flexibility of alkynes.
Cyclophane 7 has a helical chiral nature which is revealed
in its ORTEP representation and in the partial view of the
crystal-packing structure (Figures 1 and 2).15 Each pair of like-
handed helices stack along the z-axis in a herringbone manner,
but the like-handed helices are separated from the opposite-
handed helices by a channel of ODCBs. Thus an interesting
aspect of the crystal is that each helix appears to recognize its
like-handed partner. An intriguing possibility with this system
is that derivatization at the alkene positions with an appropriate
long-chain alkyl group will support the formation of chiral
columnar arrangements giving rise to interesting liquid crystal-
line properties. Another aspect of the crystallization process
that intrigued us was that of the many solvent systems used,
only ODCB produced well-formed crystals so far. The reason
for this presumably lies in the complex cocrystallization scheme
that can be seen from the arrangement of ODCBs in the lattice.
Although one ODCB is included in the space between the like-
handed cyclophanes, the middle two form a channel which
extends along the z-axis throughout the crystal lattice.
Figure 3. MALDI-TOF negative ion mass spectrum of 7.
symmetry.18 Using semiempirical and molecular mechanics
methods, the energy of the transition state was located 3-6 kcal/
mol higher in energy than that of the minimized, relaxed
enantiomeric forms. Thus even at low temperatures, the
racemization of the flexible cyclophane must occur very rapidly.
An important question to answer is whether the cyclophane
7 can be induced to rearrange to C60 in the gas phase. MALDI-
FTMS and APCI negative ion mass spectrometric studies on 7
have revealed a cluster of peaks around the molecular ion C60H18
at m/z 738 (Figure 3). The most abundant ion corresponds to
the parent ion 7, with the higher m/z peaks in the approximate
relative abundance expected for natural 13C incorporation in
C60H18 and subsequent dehydrogenated species. The lower mass
peaks from m/z 734 to 737 indicate that 7 undergoes dehydro-
genation at least down to C60H14; however no peak for C60 was
detected in the experimental conditions examined so far. The
reluctance of 7 to lose hydrogen seems at first surprising, since
its thermal chemistry must be dictated by its enediyne nature
giving rise to aromatic 1,4-diyls Via Bergman cyclization.19
However, this system may be somewhat too flexible to produce
controlled pathways for bond formation. Further studies will
be directed at the preparation of halogenated derivatives of 7
capable of generating the more rigid precursor 1.
Acknowledgment. We thank Prof. Ken Houk at UCLA for help in
the T. S. calculations. We are grateful to the Office of Naval Research
(N00014-94-1-0534) for financial support. C.L.H. acknowledges
support through a National Research Council-Naval Research Labora-
tory Research Associateship.
Supporting Information Available: Experimental procedures, 1H
and 13C NMR spectra for 4b, 6a, and 7, X-ray structure of 7, fully
labeled view of the structure, tables of atomic coordinates, equivalent
isotropic thermal parameters, bond angles, and bond lengths (22 pages).
Ordering information is given on any current masthead page.
We have determined by computation whether cyclophane 7
should be separable into its enantiomeric forms. Racemization
of the enantiomeric forms should proceed Via the dynamic
process in Table 1 resulting from rotating the two arene planes
in opposite directions, producing a transition state with D3h
JA9606384
(16) SPARTAN version 4.0.1; Wavefunction Inc., 18401 Von Karman
Avenue, Suite 370, Irvine, CA 92715.
(17) Biosym Insight II molecular modeling package.
(18) AM1 analysis of the vibrational modes produced a D3h geometry
with one imaginery frequency.
(15) For chiral cyclophanes, see: Forman, J. E.; Barrans, R. E.;
Dougherty, D. A. J. Am. Chem. Soc. 1995, 117, 9213-9228 and references
therein.
(19) Wang, K. K. Chem. ReV. 1996, 96, 207-222.