The chemical retro-Bingel reaction: Selective removal of bis(alkoxycarbonyl)methano addends from C60 and C70 with amalgamated magnesium

Article abstract of DOI:10.1039/a909704j

Bis(alkoxycarbonyl)methano addends are removed from C₆₀ and C₇₀ derivatives by reaction with amalgamated magnesium in dry THF; this facile and selective retro-Bingel reaction, which leaves pyrrolidine rings fused to C₆₀ intact, opens up the possibility of using bis(alkoxycarbonyl)methano addends as protecting and reversible directing groups in the regioselective multiple functionalization of fullerenes.

Full text of DOI:10.1039/a909704j

The chemical retro-Bingel reaction: selective removal of
bis(alkoxycarbonyl)methano addends from C60 and C70 with amalgamated
magnesium
Nicolle N. P. Moonen,a Carlo Thilgen,a Luis Echegoyenb and François Diederich*a
a Laboratorium fu¨r Organische Chemie, ETH-Zentrum, Universita¨tstrasse 16, CH-8092 Zu¨rich, Switzerland.
E-mail: diederich@org.chem.ethz.ch
b Department of Chemistry, University of Miami, Coral Gables, FL 33124, USA
Received (in Cambridge, UK) 9th December 1999, Accepted 25th January 2000
Bis(alkoxycarbonyl)methano addends are removed from C60
and C70 derivatives by reaction with amalgamated magne-
sium in dry THF; this facile and selective retro-Bingel
reaction, which leaves pyrrolidine rings fused to C60 intact,
opens up the possibility of using bis(alkoxycarbonyl)me-
thano addends as protecting and reversible directing groups
in the regioselective multiple functionalization of full-
erenes.
CuCl2 did not lead to better or more reproducible yields. The
best and most consistent results were obtained by amalgamation
of the Mg powder with 10% HgBr2 in THF, followed by a
replacement of the solvent with fresh THF, addition of the
Regioselective multiple functionalization of fullerenes is a
subject of great interest. A major strategy used to accomplish
this task is tether-directed remote functionalization.1 Another
protocol, often used in combination with tether-directed remote
functionalization,2 is the temporary introduction of an addend
which protects specific positions and, at the same time, directs
new incoming addends regioselectively to other bonds on the
carbon sphere. Examples for such addends, which are readily
introduced and removed, are the [9,10]anthraceno bridge3 as
well as fullerene-fused cyclohexene2,4 and isoxazoline5 rings.
One of the most common fullerene functionalization methods
is the nucleophilic cyclopropanation with 2-halomalonates, the
so-called Bingel reaction.6 Although the resulting bis(alk-
oxycarbonyl)methanofullerenes are generally quite stable, there
were early hints that the Bingel-type addends may be removed
under certain circumstances.7 Closer investigations led to the
discovery of the electrochemical retro-Bingel reaction, consist-
ing of the preparative removal of bis(alkoxycarbonyl)methano
addends from fullerene adducts by exhaustive electrochemical
reduction.8 The first successful applications of the Bingel–
electrochemical retro-Bingel reaction sequence were reported
in the chemistry of the higher fullerenes, with the preparation of
enantiomerically pure D2-C76 and D2-C84,9a pure D2d-C84
8
9a
and a new C84 isomer,9a and C2v-symmetric C78.9b The potential
of bis(alkoxycarbonyl)methano addends as general protecting
and reversible directing groups in fullerene chemistry prompted
us to investigate the possibility of carrying out the retro-Bingel
reaction under chemical conditions. Such a method would have
the advantages of not requiring electrochemical equipment,
avoiding the use of supporting electrolyte, and allowing
conversions on a larger scale.
Initial attempts to remove the Bingel-type addend from 1 by
electron transfer from Na metal or sodium naphthalenide in
THF failed, leading only to materials that were insoluble in THF
or toluene. Another convenient reducing agent which is easy to
handle is metallic magnesium. Heating 1 with Mg powder (25
equiv.) in dry refluxing THF under Ar led to the formation of a
brownish precipitate after ca. 1 d, and after 3 d, varying yields
of up to 81% of C60 could be isolated from the mixture. In
analogy to Grignard reactions, starting the retro-Bingel reaction
with pure Mg sometimes proved to be quite difficult. This
situation could only be partially improved through activation of
the metal by treatment with I2 prior to the addition of the
fullerene derivative. Use of the highly active Rieke magne-
sium10 only led to products that were insoluble in THF or
toluene. Also, treatment of the commercial magnesium with
Scheme 1 Reagents and conditions: i, Mg/Hg, THF, Ar, 80 °C, 3 d.
DOI: 10.1039/a909704j
Chem. Commun., 2000, 335–336
This journal is © The Royal Society of Chemistry 2000
335

View Article Online
which the solvent was replaced by fresh dry THF (8 ml) and C60
monoadduct 1 (20 mg, 23 mmol) was added. The mixture was sonicated for
another 30 min and then heated to reflux for 3 d. Column chromatography
methanofullerene and heating to reflux for 3 d.† Under these
conditions, 1 afforded C60 in 73% yield while unchanged
starting material was recovered in 23% yield (Scheme 1).
Similarly, the C70 mono-adduct 2 provided 63% of the parent
fullerene. These yields are comparable to those obtained by the
electrochemical retro-Bingel reaction.8 Traces of as yet uni-
dentified by-products, which may provide useful information
with regard to the reaction mechanism, were detected during
chromatographic purification of the products.
(SiO2; toluene–hexane 1+1) provided 12 mg (73%) of C60
.
1 L. Isaacs, R. F. Haldimann and F. Diederich, Angew. Chem., Int. Ed.
Engl., 1994, 33, 2339; F. Diederich and R. Kessinger, Acc. Chem. Res.,
1999, 32, 537; F. Diederich and R. Kessinger, in Templated Organic
Synthesis, ed. F. Diederich and P. J. Stang, Wiley-VCH, Weinheim,
1999, pp. 189–218.
2 F. Cardullo, L. Isaacs, F. Diederich, J.-P. Gisselbrecht, C. Boudon and
M. Gross, Chem. Commun., 1996, 797; W. Qian and Y. Rubin, Angew.
Chem., Int. Ed., 1999, 38, 2356.
Zn dust and the Zn/Cu couple11 are both active reducing
agents in the retro-Bingel reaction of 1, but the yields (Zn: 47%
C60, Zn/Cu: 29% C60) are lower and, particularly with pure Zn,
the amount of unidentified by-products is higher.
3 R. Schwenninger, T. Mu¨ller and B. Kra¨utler, J. Am. Chem. Soc., 1997,
119, 9317; I. Lamparth, C. Maichle-Mo¨ssmer and A. Hirsch, Angew.
Chem., Int. Ed. Engl., 1995, 34, 1607.
In the case of Bingel-type bis-adducts of C60, the yields were
variable (Scheme 1). In different experiments, the reduction of
the pure trans-3 and e regioisomers (±)-3 and 4, or of a mixture
of the seven known regioisomers,12 with Mg (100 equiv.)
amalgamated with 10% HgBr2 afforded between 48% C60 (and
13% mono-adduct 1) and 13% C60 (and 44% 1). These
fluctuations contrast the clean conversion (up to 75% yield) of
the bis-adducts to C60 by the electrochemical retro-Bingel
protocol,8 but the reasons for these differences are not yet
understood. Interestingly, whereas the product mixture re-
covered from the non-exhaustive electrochemical reduction of
constitutionally pure bis-adducts contained regioisomers of the
starting material resulting from an intramolecular ‘walk-on-the-
sphere’ rearrangement,13 such isomerization was not detected in
the reduction of pure (±)-3 or 4 with amalgamated Mg.
Fused pyrrolidines introduced by 1,3-dipolar cycloaddition
are among the most versatile addends in fullerene chemistry.
We therefore explored whether a bis(dialkoxycarbonyl)me-
thano addend could be removed from bis-adducts such as
(±)-514 while leaving the methaniminomethano bridge intact.
When (±)-515 was reacted for 3 d with amalgamated Mg (50
equiv.), fulleropyrrolidine 616 was isolated in 64% yield,
showing that the heterocycle is stable under the conditions of
the chemical retro-Bingel reaction. This result opens up the
possibility of using Bingel-type addends as protecting and
reversible directing groups in future syntheses of fullerene
multi-adducts that are not otherwise accessible. Investigations
along these lines are now under way.
4 Y.-Z. An, G. A. Ellis, A. L. Viado and Y. Rubin, J. Org. Chem., 1995,
60, 6353.
5 T. Da Ros, M. Prato, F. Novello, M. Maggini, M. De Amici and C. De
Micheli, Chem. Commun., 1997, 59.
6 C. Bingel, Chem. Ber., 1993, 126, 1957.
7 A. Hirsch, I. Lamparth, T. Gro¨sser and H. R. Karfunkel, J. Am. Chem.
Soc., 1994, 116, 9385; M. Keshavarz-K., B. Knight, R. C. Haddon and
F. Wudl, Tetrahedron, 1996, 52, 5149; F. Arias, Y. Yang, L.
Echegoyen, Q. Lu and S. R. Wilson, in Recent Advances in the
Chemistry and Physics of Fullerenes and Related Materials, ed. K. M.
Kadish and R. S. Ruoff, The Electrochemical Society, Pennington, NJ,
1995, pp. 200–212; F. Cardullo, P. Seiler, L. Isaacs, J.-F. Nierengarten,
R. F. Haldimann, F. Diederich, T. Mordasini-Denti, W. Thiel, C.
Boudon, J.-P. Gisselbrecht and M. Gross, Helv. Chim. Acta, 1997, 80,
343.
8 R. Kessinger, J. Crassous, A. Herrmann, M. Ru¨ttimann, L. Echegoyen
and F. Diederich, Angew. Chem., Int. Ed., 1998, 110, 1919.
9 (a) J. Crassous, J. Rivera, N. S. Fender, L. Shu, L. Echegoyen, C.
Thilgen, A. Herrmann and F. Diederich, Angew. Chem., Int. Ed., 1999,
38, 1613; (b) C. Boudon, J.-P. Gisselbrecht, M. Gross, A. Herrmann, M.
Ru¨ttimann, J. Crassous, F. Cardullo, L. Echegoyen and F. Diederich,
J. Am. Chem. Soc., 1998, 120, 7860.
10 R. D. Rieke and S. E. Bales, J. Am. Chem. Soc., 1974, 96, 1775; R. D.
Rieke, Acc. Chem. Res., 1977, 10, 301.
11 The Zn/Cu couple was prepared according to: H. P. Spielmann, G.-W.
Wang, M. S. Meier and B. R. Weedon, J. Org. Chem., 1998, 63,
9865.
12 A. Hirsch, I. Lamparth and H. R. Karfunkel, Angew. Chem., Int. Ed.
Engl., 1994, 33, 437.
13 R. Kessinger, M. Go´mez-Lopez, C. Boudon, J.-P. Gisselbrecht, M.
Gross, L. Echegoyen and F. Diederich, J. Am. Chem. Soc., 1998, 120,
8545.
Support from the Swiss National Science Foundation is
gratefully acknowledged.
14 M. Prato and M. Maggini, Acc. Chem. Res., 1998, 31, 519.
15 R. Kessinger and F. Diederich, unpublished results.
16 M. Maggini, G. Scorrano and M. Prato, J. Am. Chem. Soc., 1993, 115,
9798.
Notes and references
† In a typical experiment, a mixture of Mg powder (50–150 mesh, purity
!99.8%; 14 mg, 576 mmol) and HgBr2 (20 mg, 57 mmol) (both reagents
from Fluka) was sonicated in dry THF (8 ml) for 30 min under Ar, after
Communication a909704j
336
Chem. Commun., 2000, 335–336