Radical cyclization/fragmentation reactions of dicyanocyclopropanes to enaminonitriles. A radical alternative to the Thorpe-Ziegler reaction

Article abstract of DOI:10.1055/s-1999-2551

Radical cyclizations to dicyanocyclopropanes provide cyclic enaminonitriles by a series of radical reactions that features radical cyclization to a nitrile and cyclopropane cleavage. The direction of cyclopropane cleavage - and hence the ring size of the final product - depends on the nature of the ring formed in the initial cyclization to the nitrile.

Full text of DOI:10.1055/s-1999-2551

LETTER
117
Radical Cyclization/Fragmentation Reactions of Dicyanocyclopropanes to
Enaminonitriles. A Radical Alternative to the Thorpe-Ziegler Reaction
Dennis P. Curran* and Weidong Liu
Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania, 15260, USA
Received 22 September 1998
and 6-endo or 5-exo homolytic substitution reactions
(Scheme 1). This system was viewed as the most favor-
able for homolytic substitution relative to nitrile cycliza-
tion. The synthesis of 1 follows a straightforward series of
Abstract: Radical cyclizations to dicyanocyclopropanes provide
cyclic enaminonitriles by a series of radical reactions that features
radical cyclization to a nitrile and cyclopropane cleavage. The di-
rection of cyclopropane cleavage—and hence the ring size of the fi-
nal product—depends on the nature of the ring formed in the initial steps from the TBS-ether of hex-5-en-1-ol and bromoma-
cyclization to the nitrile.
lonodinitrile that features an atom transfer addition/1,3-
elimination reaction6 to form the cyclopropane.
Keywords: radical cascade, nitrile, Thorpe-Ziegler, ring expansion,
cyclopropane
The qualitative analogy between reactions of double
bonds to those of cyclopropanes is well established in
many areas of organic chemistry.1 However, in radical
chemistry, additions to carbon-carbon double bonds are
widespread2 while analogous homolytic substitution reac-
tions of cyclopropanes are rare.3 In an attempt to discover
substituent effects that would promote homolytic substi-
tution, we studied the radical reactions of a series of dicy-
anocyclopropanes (Figure 1). However, in no case was
homolytic substitution observed. Instead, the reactions
evolve through initial cyclization to one of the nitriles2b,4
followed by a series of steps that ultimately provides
enaminonitriles. This newly discovered radical sequence
complements the traditional method to make enamino-
nitriles—the Thorpe-Ziegler reaction5—both by starting
from different types of precursors and by providing access
to products that are difficult or impossible to make by the
Scheme 1
traditional route.
Cyclization of 1 with 1.5 equiv of tributyltin hydride at
both 0.1 and 0.01M provided only the product 2 of simple
reductive debromination. However, syringe pump addi-
tion of tin hydride provided 2 (33%) along with the enam-
inonitrile 3 (43%) and some starting material. The
mechanism for formation of 37,8 (see below) suggested
that 2 equiv of tin hydride was required for complete reac-
tion, and the reaction was repeated with slower syringe
pumping and now using 2.5 equiv of tin hydride. This
gave mainly 3 alongside a trace amount of 2; product 3
was isolated in 56% yield after flash chromatography. In
no experiment did we see evidence for any of the possible
homolytic substitution products.
To rationalize the formation of 3, we suggest that (Figure
2) initial radical 4 undergoes a slow (< 104 M–1s–1) cy-
clization to one of the nitriles (probably the one which is
Figure 1. Radical Cyclizations of Dicyanocyclopropanes
cis-disposed) in preference to either homolytic substitu-
tion option. Reduction of 5 by tin hydride provides an imi-
ne 6 and consumes the first equiv of tin hydride. This
We first investigated the radical cyclization of bromodini-
trile 1, which has options for 7-endo cyclization to a nitrile
Synlett 1999, No. 1, 117–119 ISSN 0936-5214 © Thieme Stuttgart · New York

118
D. P. Curran, W. Liu
LETTER
imine 6 is then subject to addition of the tin radical, cyclo-
propane cleavage and hydrogen transfer in a sequence (6
Æ 7 Æ 8 Æ 9) that consumes the second equivalent of tin
hydride. Hydrolytic workup then provides the enaminoni-
trile 3. A number of features of this cascade merit ampli-
fication. Cyclizations to nitriles are relatively slow and
can be reversible.2b,4 In this case, the rigid cyclopropane
ring must facilitate the 7-exo cyclization of 4. Fragmenta-
tion of the iminyl radical 5 does not occur in preference to
reduction to 6 since only the starting primary radical or a
cyano-substituted cyclopropyl radical could be formed.
The sequence of 6 ÆÆ 9 appears to be new, but it has pre-
cedent in related reactions of ketocyclopropanes and relat-
ed molecules.9,10 Cleavage of the transannular bond is the
expected outcome with a cyclopropyl-fused cyclohexyl
radical, although the results do not exclude reversible
cleavage of the lateral bond.8 The overall transformation
in Scheme 1 represents a straightforward route from an
acyclic precursor to a functionalized cyclooctane ring, and
it features radical reactions for all the key C–C bond form-
ing steps.
Scheme 2
Although we have only studied a limited number of exam-
ples, the synthetic method introduced in this paper shows
good promise for application in diverse and complex set-
tings. The individual steps in the sequence can each be re-
lated back to known reactions and their success should be
relatively predictable in other analogs. In essence, the
method is a strategic alternative to the classical Thorpe-
Ziegler condensation of dinitriles, which is the most pop-
ular way to make enaminonitriles. However, the radical
approach allows access to otherwise difficult or impossi-
ble products. For example, cyclooctanes are not easy to
form by the Thorpe-Ziegler condensation. And substitu-
tion patterns like 11a-c are difficult to form regioselec-
tively in the traditional Thorpe-Ziegler condensation due
to problems with regioisomers. Facile access to these
types of enaminonitriles should expand the synthetic util-
ity of this class of molecules.
Acknowledgement
Figure 2. Suggested Mechanism for Formation of Enaminonitrile 3
We thank the National Science Foundation for funding of this work.
We briefly investigated the scope of this transformation
by studying the reactions of cyclopropane dinitriles 10a-d
(Scheme 2). The syntheses of these substrates are not
shown; they all started with the corresponding TBS-alk-
enol and followed an identical sequence of steps to that
shown in Scheme 1. The halides were reduced by the sy-
ringe pump procedure with 2.5 equiv of tin hydride and
then the products were purified by flash chromatography.
Not surprisingly, the substrate 10d capable of 8-exo cy-
clization provided only the product 12d of reductive de-
bromination. In contrast, substrates capable of 5-exo
(10a,b) or 6-exo (10c) cyclization to the nitrile underwent
a series of reactions similar to that in Figure 2, but cleaved
the lateral bond rather than the transannular bond to pro-
vide methyl substituted enaminonitriles 11a-c. This stere-
oelectronically controlled mode of cleavage is anticipated
based on prior results of structurally related radicals.8
References and Notes
(1) Wong, H. N. C. In Rodd’s Chemistry of Carbon Compounds,
Vol. 2. Alicyclic Compounds.; M. Sainsbury, Ed.; Elsevier:
Amsterdam, The Netherlands, 1994.
(2) a) Curran, D. P. In Comprehensive Organic Synthesis; B. M.
Trost and I. Fleming, Ed.; Pergamon: Oxford, 1991; Vol. 4; pp
715. b) Curran, D. P. In Comprehensive Organic Synthesis; B.
M. Trost and I. Fleming, Ed.; Pergamon: Oxford, 1991; Vol.
4; pp 779.
(3) a) Recent review of homolytic substitutions: Schiesser, C. H.;
Wild, L. M. Tetrahedron 1996, 53, 13265. b) Wiberg, K. B.;
Waddell, S. T.; Laidig, K. Tetrahedron Lett. 1986, 27, 1553;
c) Banz, U.; Polborn, K.; Wagner, H.-U.; Szeimies, G. Chem.
Ber. 1988, 121, 1785. d) Kaszynski, P.; Michl, J. J. Am. Chem.
Soc. 1988, 110, 5225.
(4) Giese, B.; Kopping, B.; Dickhaut, J.; Thomas, G.; Kulicke, K.
J. Org. React. 1996, 48, 315.
Synlett 1999, No. 1, 117–119 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Radical Cyclization/Fragmentation Reactions of Dicyanocyclopropanes to Enaminonitriles
119
(5) a) Taylor, E. C.; Alexander, M. The Chemistry of Cyclic Ena-
minonitriles and o-Aminonitriles; Wiley, New York, 1970 b)
Lydia, R. H.; Margarita, P. M.; Mariano, M. Synth. Commun.
1984, 14, 967.
(6) a) Peter, B.; Lothar, S.; Jutta, E. Chem. Ber. 1967, 100, 1281.
b) Curran, D. P.; Seong, C. M. Tetrahedron 1992, 48, 2157. c)
Curran, D. P.; Seong, C. M. Tetrahedron 1992, 48, 2175.
(7) Karle, D. W. Ph.D. Thesis, 1965, University of Southern Ca-
lifornia, Los Angeles, CA.
(11b): 1H NMR d 4.32 (bs, 2 H), 2.48 (t, J = 7, 2 H), 1.77 (t, J
= 7, 2 H), 1.18 (s, 6 H); 13C NMR d 159.81, 118.14, 86.15,
44.73, 37.91, 32.71, 28.24; IR 3442, 3299, 2950, 2192, 1663,
1345, 1610; MS m/e 136, 121, 106, 94, 77, 67. 2-Amino-6-
methylcyclohex-1-enecarbonitrile (11c): 1H NMR d 4.22 (bs,
2 H), 2.37 (m, 1 H), 2.14 (m, 2 H), 1.79 (m, 2 H), 1.57 (m, 2
H), 1.18 (d, J = 3, 3 H); 13C NMR d 156.00, 120.31, 81.13,
30.68, 28.95, 28.47, 21.13, 19.90; IR; 3368, 2961, 2182, 1722,
1643, 1466; MS m/e 136, 122, 108, 94, 81, 67, 54.
(8) 2-Aminocyclooct-1-enecarbonitrile (3): 1H NMR d 4.33 (bs, 2
H), 2.28 (t, J = 6, 2 H), 2.22 (t, J = 6, 2 H), 1.48-1.67(m, 8 H);
13C NMR d 160.13, 121.54, 75.28, 31.02, 30.31, 28.94, 26.98,
25.95, 25.72; IR 3427, 3350, 3245, 2931, 2854, 2324, 2175,
1638, 1600, 1442, 1194, 1055; MS m/e 150, 122, 107, 94, 82,
67, 54. 2-Amino-5-methylcyclopent-1-enecarbonitrile (11a):
1H NMR d 4.50 (bs, 2 H), 2.91 (m, 1 H), 2.44 (t, J = 7, 2 H),
2.13 (m, 1 H), 2.13 (d J = 7, 3 H), 1.42 (m, 1 H); 13C NMR d
151.17, 118.71, 81.09, 38.85, 33.55, 31.05, 20.77; IR 3486,
3376, 2979, 2191, 1669, 1396, 1616; MS m/e 122, 107, 94, 80,
67, 53. 2-Amino-5,5-dimethylcyclopent-1-enecarbonitrile
(9) Additions of tin radicals to C–N multiple bonds: Fallis, A. G.;
Brinza, I. M. Tetrahedron 1997, 53, 17543.
(10) Relevant cyclopropylcarbinyl radical chemistry: a) Nonhebel,
D. C. Chem. Soc. Rev. 1993, 22, 347. b) Dowd, P.; Zhang, W.
Chem. Rev. 1993, 93, 2091. c) Newcomb, M. Tetrahedron
1993, 49, 1151. d) Ingold, K. U.; Walton, J. C. Acc. Chem.
Res. 1986, 19, 72. e) Beckwith, A. L. J.; Moad, G. J. Chem.
Soc., Perkin Trans. 1980, 2, 1473. f) Clive, D. L. J.; Daigne-
ault, S. J. Org. Chem. 1991, 56, 3801. g) Batey, R. A.; Harling,
J. D.; Motherwell, W. B. Tetrahedron 1996, 52, 11421.
Synlett 1999, No. 1, 117–119 ISSN 0936-5214 © Thieme Stuttgart · New York