Generation and intermolecular capture of cyclopropylacyl radicals

Article abstract of DOI:10.1021/ol047892i

(Chemical Equation Presented) Cyclopropylacyl radicals derived from S-cyclopropylacyl xanthates (dithiocarbonates) undergo intermolecular additions to olefins without loss of CO or ring opening. In the presence of a phenyl ring on carbon C-1 of the cyclopropane ring, loss can be made to occur in the absence of an olefinic trap. The adducts from the cyclopropylacyl radical additions are easily converted into enones by base-induced β-elimination of the xanthate group.

Full text of DOI:10.1021/ol047892i

ORGANIC
LETTERS
2004
Vol. 6, No. 26
4969-4972
Generation and Intermolecular Capture
of Cyclopropylacyl Radicals
Markus R. Heinrich and Samir Z. Zard*
Laboratoire de Synthe`se Organique associe´ au CNRS, Ecole Polytechnique,
91128 Palaiseau Cedex, France
zard@poly.polytechnique.fr
Received October 12, 2004
ABSTRACT
Cyclopropylacyl radicals derived from S-cyclopropylacyl xanthates (dithiocarbonates) undergo intermolecular additions to olefins without loss
of CO or ring opening. In the presence of a phenyl ring on carbon C-1 of the cyclopropane ring, loss can be made to occur in the absence
of an olefinic trap. The adducts from the cyclopropylacyl radical additions are easily converted into enones by base-induced
the xanthate group.
â-elimination of
As an illustration of the synthetic potential of the radical
transfer of xanthates and related groups,1 we contemplated
the possibility of constructing ꢀ-lactam 1, the core structure
of various vasopeptidase inhibitors of some recent impor-
tance,2 by the addition of xanthate 3 to protected allyl glycine
4 as outlined in Scheme 1. The acyl radical 5 arising from
3 would be expected to rapidly lose carbon monoxide to give
the stabilized tertiary radical 6, which would then undergo
the desired addition to the olefinic trap to give ultimately
adduct 8.3 This compound has two differentially protected
amines and all the elements necessary to constitute an
immediate precursor to the target molecule 1. The synthesis
of the requisite xanthate 3 was straightforward from the
known acid chloride 2,4 and conditions were found to induce
its addition to simple olefins such as allyl acetate and vinyl
pivalate, affording adducts 9 and 10 in 76 and 94% yields,
respectively (Scheme 1). However, the addition proceeded
quite poorly with protected allyl glycine 4. Only a low yield
(<10%) of desired adduct 8 could be secured by heating at
60 °C in 1,2-dichloroethane in the presence of lauroyl
peroxide as an initiator.
The main problem was the tendency of xanthate 11, which
is in equilibrium with radical 6, to undergo elimination of
xanthic acid 12 to give olefin 13.5 The xanthic acid released
in this process acts as an inhibitor for the radical chain. This
unimolecular elimination is favored by an increase in
temperature, hence the need to operate below 60 °C, in
contrast to the usual radical additions we have routinely
performed over the years in temperature ranges of 80-130
°C.
(1) For reviews, see: Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997,
36, 672-685; Angew. Chem. 1997, 109, 724-737.
(2) (a) Singh, J.; Kronenthal, D. R.; Schwinden, M.; Godfrey, J. D.; Fox,
R.; Vawter, E. J.; Zhang, B.; Kissick, T. P.; Patel, B.; Mneimne, O.; Humora,
M.; Papaioannou, C. G.; Szymanski, W.; Wong, M. K. Y.; Chen, C. K.;
Heikes, J. E.; DiMarco, J. D.; Qiu, J.; Deshpande, R. P.; Gougoutas, J. Z.;
Mueller, R. H. Org. Lett. 2003, 5, 3155-3158. (b) Robl, J. A.; Sieber-
McMaster, E.; Sulsky, R. Tetrahedron Lett. 1996, 37, 8985-8988.
(3) (a) Delduc, P.; Tailhan, C.; Zard, S. Z. J. Chem. Soc., Chem. Commun.
1988, 308-310. (b) For a review of acyl radicals, see: Ryu, I.; Sonoda,
N.; Curran, D. P. Chem. ReV. 1996, 96, 177-194. (c) For more recent work,
see: Bath, S.; Lopez-Ruiz, H.; Laso, N. M.; Quiclet-Sire, B.; Zard, S. Z.
Chem. Commun. 2003, 204-205. (d) Tsujimoto, S.; Sakaguchi, S.; Ishii,
Y. Tetrahedron Lett. 2003, 44, 5601-5604. (e) Tsujimoto, S.; Iwahama,
T.; Sakaguchi, S.; Ishii, Y. Chem. Commun. 2001, 2352-2353. (f) Roberts,
B. P.; Winter, J. N. Chem. Soc. ReV. 1999, 25-35.
To limit this problem, we examined the use of the
cyclopropyl analogue 15, where thermal elimination of the
xanthate group would be expected to be difficult due to the
strain inherent in the resulting olefin. At the same time, this
would allow access to perhaps even more interesting vaso-
(4) For preparation of acid chloride 2, see: Aitken, R. A.; Cooper, H.
R.; Mehrotra, A. P. J. Chem. Soc., Perkin Trans. 1 1996, 475-484.
(5) We have previously observed a similar thermal elimination of a
xanthate group from adducts with N-vinyl pyrrolidine: Gagosz, F.; Zard,
S. Z. Org. Lett. 2003, 5, 2655-2657.
10.1021/ol047892i CCC: $27.50
© 2004 American Chemical Society
Published on Web 12/02/2004

Scheme 1
Scheme 2
destabilizing effect of the cyclopropyl ring. Indeed, irradia-
tion of xanthate 15 in refluxing 1,2-dichloroethane for several
hours did not cause any decomposition and the starting
material was recovered essentially unchanged in >95% yield.
In the absence of the olefinic trap and with no possibility of
losing carbon monoxide, the cyclopropylacyl radical 16 is
condemned to evolve back into the starting material by
reacting with its precursor, xanthate 15.
The addition to allyl acetate described above appears to
be the first case where a cyclopropylacyl radical is generated
and captured in an intermolecular fashion. We are aware of
only one example reported by Pattenden and co-workers,8d
where a cyclopropylacyl radical was intercepted by an
internal olefin. Interestingly, the cyclopropylacyl radical does
not undergo ring opening to ketene 21 in the present case.
Scission to the ketene was observed earlier by Pattenden’s
group,8d but only when the ensuing radical was stabilized
by conjugation to a carbonyl group.
As shown by the transformations depicted in Scheme 3,
additions to various olefins could be accomplished in
synthetically useful yields. Except for vinyl pivalate, the use
of peroxide proved to be more efficient than photochemical
initiation.9 We also found it beneficial to add most of the
peroxide at the beginning in order to favor the radical chain
addition over the slow decomposition of the starting xanthate.
peptidase inhibitors derived from 19, where the geminal
dimethyl group has been replaced by a metabolically more
robust cyclopropane (Scheme 2).2b The desired acyl xanthate
15 was readily prepared by reaction of potassium O-ethyl
xanthate with the corresponding acid chloride 14.6 When this
new xanthate was irradiated in the presence of allyl acetate
in refluxing 1,2-dichloroethane, we were surprised to find
that adduct 20, isolated in 46% yield, still retained the
carbonyl group. An improved yield (63%) was obtained when
initiation was conducted with 10 mol % lauroyl peroxide
instead of visible light.
Clearly, the cyclopropylacyl radical intermediate 16 did
not readily undergo decarbonylation to give 17, in contrast
to the dimethyl analogue. Cyclopropyl radicals resemble to
a certain extent vinyl radicals.7 They are less stable than their
aliphatic counterparts, and this is apparently sufficient to slow
considerably the extrusion of carbon monoxide.8 Nor does
the lone pair of the nitrogen seem to counterbalance the
(8) Stability of cyclopropylacyl radicals: (a) Kerr, J. A.; Smith, A.;
Trotman-Dickenson, A. F. J. Chem. Soc. A 1969, 1400-1403. (b) Blum,
P. M.; Davies, A. G.; Sutcliffe, R. J. Chem. Soc., Chem. Commun. 1979,
217-218. (c) Davies, A. G.; Sutcliffe, R. J. Chem. Soc., Perkin Trans. 2
1982, 1483-1488. (d) De Boeck, B.; Herbert, N.; Pattenden, G. Tetrahedron
Lett. 1998, 39, 6971-6974.
(6) For preparation of acid chloride 14, see: Haddow, J.; Suckling, C.
J.; Wood, H. C. S. J. Chem. Soc., Perkin Trans. 1 1989, 1297-1304
(modified procedure).
(7) For a review of cyclopropyl radicals, see: Walborsky, H. M.
Tetrahedron 1981, 37, 1625-1651.
(9) Initiation with DLP led to polymerization of vinyl pivalate. A similar
result was obtained when a solution of 15 in vinyl pivalate was irradiated.
4970
Org. Lett., Vol. 6, No. 26, 2004

Scheme 3
Scheme 4
Acyl xanthates in general are hydrolytically labile,1 and
degradation by this pathway liberates xanthic acid and other
sulfur-containing side products that, as was stated above, act
as inhibitors. None of the additions presented in Scheme 3
have been optimized, and room for improvement certainly
exists. Nevertheless, the process provides access to structures
with a dense combination of functionalities that would be
very difficult to obtain otherwise. The addition to protected
allyl glycine 4 is especially noteworthy since it opens a route
to unusual amino acids. The positioning of the xanthate group
â- to the carbonyl group in the various adducts allows its
elimination with base to produce the unsaturated ketones.
In the case of 24, the resulting olefin can migrate along the
chain by a series of prototropic shifts to give ultimately the
thermodynamically more stable enamide 27. Such enamides
are very valuable substrates since it is well established that
their hydrogenation in the presence of chiral catalysts can
give rise to amino acids with high optical purity.10
aldehyde, so that exposure to hydrazine resulted in the
formation of 3-cyclopropyl-pyrazole 32 in 60% yield. The
volatility of this compound made its isolation on a small
scale somewhat inconvenient. The more complex xanthate
33, a compound readily prepared from (+)-2-carene,12 also
participated in the radical chain addition as illustrated by its
addition to trimethylvinylsilane to give adduct 34 in 42%
yield. Base-induced elimination of the xanthate group
Scheme 5
The reaction could be extended to other cyclopropylacyl
xanthates (Scheme 4). The simplest derivative, 28, is less
protected by steric hindrance against hydrolysis but neverthe-
less underwent the desired addition. Addition to octene was
immediately followed by treatment with DBU to give directly
enone 30 in 79% yield ((E)-isomer).11 In the case of the
adduct 31 arising from the reaction with vinyl pivalate, the
carbon bearing the xanthate is at the oxidation level of an
(10) Tang, W.; Zhang, X. Chem. ReV. 2003, 103, 3029-3070.
(11) For previous syntheses of enone 30, see: (a) Cheskis, B. A.; Ivanova,
N. M.; Moiseenkov, A. M.; Nefedov, O. M. Bull. Acad. Sci. USSR, DiV.
Chem. Sci. 1990, 39, 1839-1849. (b) Moiseenkov, A. M.; Cheskis, B. A.;
Ivanova, N. M.; Nefedov, O. M. J. Chem. Soc., Perkin Trans. 1 1991, 2639-
2649. (c) Cheskis, B. A.; Ivanova, N. M.; Moiseenkov, A. M.; Nefedov,
O. M. Bull. Acad. Sci. USSR, DiV. Chem. Sci. 1991, 40, 1372-1380.
Org. Lett., Vol. 6, No. 26, 2004
4971

furnished unsaturated ketone 35, again solely as the (E)-
isomer. Such vinylsilyl ketones are not trivial to make by
other routes.13 Xanthate 33 is an interesting synthon for the
enantioselective synthesis of terpenes containing a geminal
dimethylcyclopropyl motif.
mers) and presumably smaller amounts of higher oligomers.
The ready formation of such side products is again a
reflection of the relative instability of radical 38, despite the
presence of the geminal phenyl ring.
The above preliminary results provide insight into the
special behavior of cyclopropylacyl radicals. This approach
also represents an efficient and flexible route to cyclopropyl
ketones. These have proved to be exceptionally useful
substrates for organic synthesis.14
To define what group would be needed to induce loss of
carbon monoxide, we prepared and examined the phenyl-
substituted derivative 36. In this case, irradiation in refluxing
toluene in the absence of a trap resulted in decarbonylation
to give xanthate 39 in over 85% yield. Stabilization by the
phenyl group thus compensates the untoward effect of the
cyclopropyl ring and encourages the expulsion of carbon
monoxide from cyclopropylacyl radical 37. This extrusion
nevertheless remains comparatively slow since irradiation
in refluxing allyl acetate furnished ketone 40 as the major
product (61%) along with xanthate 39, which was isolated
in only 22% yield. With the more reactive and lower boiling
vinyltrimethylsilane, the reaction gave almost exclusively the
addition product 41 (87%). Treatment with DBU again
resulted in clean elimination of the xanthate group to provide
the vinylsilyl ketone 42 (71% from 36). The steric bulk of
the phenyl group in 36 protects against hydrolytic degradation
and allows efficient additions to be performed photochemi-
cally with visible light. We found that xanthate 39, arising
from the decarbonylation process, can be made to add to
allyl acetate to give 43 (22%), but this is complicated by
further addition leading to 44 (25%, mixture of diastereo-
Acknowledgment. This paper is dedicated with respect
to Professor Pierre Potier on the occasion of his 70th
birthday. We thank the Fondation Alfred Kastler for generous
financial support to M.H. and Prof. Dr. M. Spiteller
(University Dortmund) and Dr. W. Spahl (Ludwig Maxi-
milians University Munich) for HRMS analysis.
Supporting Information Available: Experimental pro-
cedures and detailed analytical data for compounds 3, 8-10,
15, 20, 22-25, 27, 28, 30-36, and 39-44. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL047892I
(14) (a) Stevens, R. V. Acc. Chem. Res. 1977, 10, 193-198. (b) For a
recent application of the Stevens reaction to the synthesis of the strychnos
alkaloids, see: Rawal, V. H.; Michoud, C.; Monestel, R. F. J. Am. Chem.
Soc. 1993, 115, 3030-3031. (c) Rawal, V. H.; Iwasa, S. J. Org. Chem.
1994, 59, 2685-2686. (d) For general reviews on activated cyclopropanes,
see: Reissig, H.-U.; Zimmer, R. Chem. ReV. 2003, 103, 1151-1196. (e)
Hudlicky, T.; Reed, J. W. In ComprehensiVe Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 5, pp 899-970. (f)
For a recent example of a thermal Cloke rearrangement of cyclopropyl
ketones and imines, see: Funke, C.; Es-Sayed, M.; de Meijere, A. Org.
Lett. 2000, 2, 4249-4251. (g) Wittig-type methylenation of cyclopropyl
ketones leads to vinylcyclopropanes, which undergo thermal rearrangement
to cyclopentenes: Baldwin, J. E. Chem. ReV. 2003, 103, 1197-1212.
(12) Donkervoort, J. G.; Gordon, A. R.; Johnstone, C.; Kerr, W. J.; Lange,
U. Tetrahedron 1996, 52, 7391-7420.
(13) (a) Silylvinyl ketones by elimination: Christiansen, M. L.; Benneche,
T.; Undheim, K. Acta Chem. Scand. Ser. B. 1987, 41, 536-540, Lee, J.;
Lee, P. H. Tetrahedron Lett. 1996, 37, 9305-9306. (b) Silylvinyl ketones
by oxidation: Denmark, S. E.; Herbert, B. J. Org. Chem. 2000, 65, 2887-
2896.
4972
Org. Lett., Vol. 6, No. 26, 2004