Three-carbon ring expansion by cyclopropane insertion: Macrocyclic musks from readily available C-12 starting materials

Article abstract of DOI:10.1021/ol0487027

(Equation Presented) A thermal three-carbon ring expansion based on side chain ring insertion of a cyclopropane moiety is described. Flash vacuum pyrolysis (FVP) of 1-cyclopropyl-cycloalk-3-enol derivatives leads to the three-carbon ring expanded enones with clean retention of double bond geometry. Substrates bearing methyl groups on the cyclopropane ring undergo regioselective bond cleavage, allowing for the systematic preparation of selectively substituted macrocyclic musks from low-priced C-12 starting compounds.

Full text of DOI:10.1021/ol0487027

ORGANIC
LETTERS
2004
Vol. 6, No. 17
2989-2991
Three-Carbon Ring Expansion by
Cyclopropane Insertion: Macrocyclic
Musks from Readily Available C-12
Starting Materials
Georg Ru1edi,* Matthias Nagel,† and Hans-Ju1rgen Hansen
Organisch-chemisches Institut der UniVersita¨t, 8057 Zu¨rich, Switzerland
georg@access.unizh.ch
Received July 7, 2004
ABSTRACT
A thermal three-carbon ring expansion based on side chain ring insertion of a cyclopropane moiety is described. Flash vacuum pyrolysis
(FVP) of 1-cyclopropyl-cycloalk-3-enol derivatives leads to the three-carbon ring expanded enones with clean retention of double bond geometry.
Substrates bearing methyl groups on the cyclopropane ring undergo regioselective bond cleavage, allowing for the systematic preparation of
selectively substituted macrocyclic musks from low-priced C-12 starting compounds.
Macrocyclic musks are of central importance in fragrance
chemistry, particularly for their excellent odor properties and
biodegradability.1 In recent years, intensive efforts have been
made to develop efficient methods that are capable of
producing substituted 15-carbon macrocyclic ketones having
an unsaturation of defined double bond geometry.2 However,
well-established macrocyclization methodologies suffer from
inherent high-dilution conditions and often produce E/Z
mixtures, whereas fragmentation procedures are limited in
the variability of appropriately substituted bicyclic precur-
sors.3 Herein, we report a fundamentally new approach based
on side chain ring insertion of a cyclopropane moiety that
allows for the systematic preparation of both specifically
substituted and specifically unsaturated C-15 macrocycles
from inexpensive C-12 starting materials in a single synthetic
operation. In addition, we demonstrate that this concept is
also applicable to other ring sizes.
Recently we reported a diradical-mediated two-carbon ring
expansion of 1-vinylcycloalkanols by means of flash vacuum
pyrolysis (FVP).4,5 The instigating question of the current
study was directed at whether the vinyl group might be
replaced with a cyclopropane ring to yield the corresponding
three-carbon ring-expanded cycloalkanones 2 by means of
a 1,4-C shift.6 The design of this proposed reaction 1 f 2
was predicated on the well-known fact that cyclopropyl-
carbinyl radicals a immediately rearrange into homoallyl
radicals b, once they have been generated (Scheme 1).7,8
† Current address: Swiss Federal Laboratories for Materials Testing and
Research (EMPA), CH-8600 Du¨bendorf, Switzerland.
(1) For leading references on musk odorants, see: (a) Fra´ter, G.;
Bajgrowicz, J. A.; Kraft, P. Tetrahedron 1998, 54, 7633. (b) Williams, A.
S. Synthesis 1999, 1707. (c) Fra´ter, G.; Bajgrowicz, J. A.; Denis, C.; Kraft,
P. Angew. Chem., Int. Ed. 2000, 39, 2980. (d) Fra´ter, G.; Lamparsky, D. In
Perfumes: Art, Science and Technology; Mu¨ller, P. M., Lamparsky, D.,
Eds.; Elsevier: London, New York, 1991; Chapter 20, pp 533-555. (e)
Ohloff, G. Riechstoffe und Geruchssinn. Die Molekulare Welt der Du¨fte;
Springer: Berlin, 1990; Chapter 9, pp 195-219. (f) Mookherjee, B. D.;
Wilson, R. A. In Fragrance Chemistry: The Science of the Sense of Smell;
Theimer, E. T., Ed.; Academic Press: New York, 1982; Chapter 12, pp
433-494.
(3) Recent examples: (a) Kamat, V. P.; Hagiwara, H.; Katsumi, T.;
Hoshi, T.; Suzuki, T.; Ando, M. Tetrahedron 2000, 56, 4397. (b) Louie, J.;
Bielawski, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2001, 123, 11312.
(4) (a) Nagel, M.; Fra´ter, G.; Hansen, H.-J. Synlett 2002, 275. (b) Nagel,
M.; Fra´ter, G.; Hansen, H.-J. Synlett 2002, 280.
(5) For reviews on FVP, see: (a) McNab, H. Contemp. Org. Synth. 1996,
3, 373. (b) Wiersum, U. E.; Jenneskens, L. W. In Gas-Phase Reactions in
Organic Synthesis; Valle´, Y., Ed.; Gordon and Breach: Amsterdam, 1997;
pp 143-194. (c) McNab, H. Aldrichimica Acta 2004, 37, 19.
(2) Fehr, C.; Galindo, J.; Etter, O.; Thommen, W. Angew. Chem., Int.
Ed. 2002, 41, 4523.
(6) Ru¨edi, G.; Hansen, H.-J. Tetrahedron Lett. 2004, 45, 5143.
10.1021/ol0487027 CCC: $27.50 © 2004 American Chemical Society
Published on Web 07/28/2004

isolated yield. It is noteworthy to state that the endocyclic
double bond suffered no E/Z isomerization by passage from
3 to 5.
Scheme 1
Interestingly, we observed a minor amount of another
isomer (6, 10%), which was ascertained to be 5-vinylcyclo-
tridecanone.13 This byproduct of mechanistic significance
presumably arises through a competing homo silyloxy-Cope
rearrangement (3,4-C shift).14
Our initial attempts to effect the diradical-mediated three-
carbon ring expansion with a series of substrates 1 of
different ring size (n ) 10-15; R ) TMS) proved
unsatisfactory. Only traces of the desired ring-expanded
products 2 could be detected when samples of 1 were
subjected to FVP (550-700 °C) followed by hydrolysis of
the intermediate silylenol ethers. Apparently, the overall
radical stabilizing ability of a cyclopropyl and a silyloxy
group at a common carbon atom is too low to effect clean
homolytic breakage.9
To reduce the activation barrier, a radical stabilizing
functionality was introduced adjacent to the bond to be
broken. Thus, an endocyclic allylic double bond gave rise
to a clearly defined weakest single bond, facilitating the
intended breakage of the C1-C2 bond.10 As a result, FVP
of 311 at 530 °C proceeded smoothly to give the ring-
expanded silylenol ether 4 as a 1:1 mixture of diastereo-
isomers (Scheme 2).12 For ease of characterization and
To study the generality of this new three-carbon ring
expansion, we examined the influence of the double bond
geometry and the ring size (Table 1, entries 1-3). FVP of
7a possessing a (Z)-double bond produced exclusively (Z)-
6-cyclopentadecenone (8a). In agreement with 3 the ho-
mologous compound 7b underwent clean homolysis to give
Table 1. Three-Carbon Ring Expansion with Various
Substrates 7a-g
Scheme 2
handling 4 was hydrolyzed by brief treatment of the crude
product with 1% HCl in EtOH to give enone 5 in 70%
(7) For a review, see: Newcomb, M. Tetrahedron 1993, 49, 1151.
(8) (a) Effio, A.; Griller, D.; Ingold, K. U.; Beckwith, A. L. J.; Serelis,
A. K. J. Am. Chem. Soc. 1980, 102, 1734. (b) Newcomb, M.; Glenn, A. G.
J. Am. Chem. Soc. 1989, 111, 275. (c) Martinez, F. N.; Schlegel, H. B.;
Newcomb, M. J. Org. Chem. 1996, 61, 8547. (d) Martinez, F. N.; Schlegel,
H. B.; Newcomb, M. J. Org. Chem. 1998, 63, 3618. (e) Halgren, T. A.;
Roberts, J. D.; Horner, J. H.; Martinez, F. N.; Tronche, C.; Newcomb, M.
J. Am. Chem. Soc. 2000, 122, 2988. (f) Choi, S.-Y.; Horner, J. H.; Newcomb,
M. J. Org. Chem. 2000, 65, 4447. (g) Horner, J. H.; Newcomb, M. J. Am.
Chem. Soc. 2001, 123, 4364. (h) Takekawa, Y.; Shishido, K. J. Org. Chem.
2001, 66, 8490. (i) Cooksy, A. L.; King, H. F.; Richardson, W. H. J. Org.
Chem. 2003, 68, 9441.
(9) (a) McMillen, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982,
33, 493. (b) Smith, D. M.; Nicolaides, A.; Golding, B. T.; Radom, L. J.
Am. Chem. Soc. 1998, 120, 10223.
(10) Brocks, J. J.; Beckhaus, H.-D.; Beckwith, A. L. J.; Ru¨chardt, C. J.
Org. Chem. 1998, 63, 1935.
(11) Compound 3 was prepared on a multigram scale by dehydrochlo-
rination of 2-chlorocyclodocecanone to give the thermodynamically more
stable 3-enone, followed by Grignard reaction and silylation (70% over
three steps).
(12) Only tar formation was observed when 3 was heated under static
conditions in a sealed glass tube at 200-400 °C.
a Overall yields for ring expansion and hydrolysis step. b GC yield
(isolated yield). c Not separated from byproducts.
2990
Org. Lett., Vol. 6, No. 17, 2004

6-cyclohexadecenone (8b) in comparable yield. As expected
from general ring strain effects in medium ring systems the
eight-membered substrate 7c was reluctant to react in this
manner. However, at more elevated temperatures (560 °C)
the 11-membered product 8c was obtained in 30% yield.
The scope was further explored with methyl-substituted
substrates 7d-g, as they would provide enones 8d-g,
representing new compounds of possible value in fragrance
chemistry. The key step for the preparation of the precursors
7d-f includes the regioselective cyclopropanation of the
exocyclic allylic double bond in dienols 9d-f (Scheme 3).
Scheme 4
Scheme 3
of the substrates 7a-g underwent E/Z isomerization during
the short thermal impact.
According to Scheme 4 the TMS-protecting group proved
to be crucial. As exemplified with cyclopropyl alcohol 10,
prolysis under comparable conditions produced three iso-
meric compounds. The major component turned out to be
ring-opened cyclopropyl ketone 11 (65%), the product of a
retro-ene fragmentation, instead of the ring expansion product
5 (5%). In addition, we found a 5,12 fused bicyclic ketone
12 (10%).19 A reasonable explanation for the formation of
the latter involves a six-electron oxa-ene process of the Conia
type within the initially formed enol form of 5.20
In summary, we have developed a new type of three-
carbon ring expansion offering a strategically novel and
practical route to an otherwise difficult-to-achieve but
common ring system. Both the low price and the easy
accessibility of C-12 precursors allow for the synthesis of
specifically substituted C-15 macrocyclic musks. Enones
8d-g are new and reveal the typical musk odor. It is of
further synthetic significance that the FVP reactions can be
run as a continuous, solvent-free process, rendering this
procedure very promising for large-scale operations. Further
studies are in progress.
As summarized in Table 1 (entries 4-7) the desired
muscenones 8d-g were obtained in good yields. The lower
yield observed in the case of 7e with a cis-substituted
cyclopropane moiety having energetically unequal bonds may
be ascribed to competing cleavage and H-transfer reactions.15
As expected, however, the more highly substituted bond
turned out to be the main site of breakage to give 4-methyl
enone 8e.16 The same result was obtained when the trans-
analogue of 7e was submitted to FVP.17 Product 8f possess-
ing a gem-dimethyl group was obtained in the same manner
from the sterically crowded precursor 7f. Finally, we
investigated the effect of a methyl group in the 2-position.
Subjecting substrate 7g to FVP at 520 °C afforded 8g bearing
the substituent at C5.18 As observed in the case of 3, neither
(13) For related silyloxy-Cope reactions, see: (a) Thies, R. W.; Billig-
meier, J. E. J. Am. Chem. Soc. 1974, 96, 200. (b) Thies, R. W.; Bolesta, R.
E. J. Org. Chem. 1976, 41, 1233. (c) Thies, R. W.; Darwula, K. P. J. Chem.
Soc., Chem. Commun. 1985, 17, 1188. (d) Thies, R. W.; Darwula, K. P. J.
Org. Chem. 1987, 52, 3798.
(14) Although many silyloxy-Cope reactions can be found in the
literature, the involvement of a cyclopropyl moiety is unprecedented so
far. For a recent review on silyloxy-Cope rearrangements, see: Schneider,
C. Synlett 2002, 1079.
Acknowledgment. This work was generously supported
by the Swiss National Science Foundation (SNF).
Supporting Information Available: Experimental condi-
tions and complete spectral data for all compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(15) Recent examples: (a) Ru¨edi, G.; Nagel, M.; Hansen, H.-J. Org.
Lett. 2003, 5, 2691. (b) Ru¨edi, G.; Nagel, M.; Hansen, H.-J. Org. Lett.
2003, 5, 4211. (c) Ru¨edi, G.; Nagel, M.; Hansen, H.-J. Synlett 2003, 1210.
(16) (a) Takekawa, Y.; Shishido, K. Tetrahedron Lett. 1999, 40, 6817.
(b) Takekawa, Y.; Shishido, K. J. Org. Chem. 2001, 66, 8490.
(17) At temperatures above 450 °C, both pure substrates interconverted
to give a 1:1 mixture of cis/trans isomers.
OL0487027
(19) The formation of the cis- as well as the trans-isomer of 12 would
generally be possible. Although only a single isomer had been formed, its
relative configuration could not be assigned by NOE measurements because
of overlapping H signals.
(20) (a) Conia, J. M.; Le Perchec, P. Synthesis 1975, 1. (b) Conia, J.
M.; Lange, G. L. J. Org. Chem. 1978, 43, 564.
1
(18) The relative configuration of compound 7g was determined to be
(1R*,2R*).
Org. Lett., Vol. 6, No. 17, 2004
2991