Mn(III)-based oxidative free-radical cyclizations of unsaturated ketones

Article abstract of DOI:10.1021/jo961199u

Mn(III)-based oxidative free-radical cyclization of unsaturated ketones is a versatile synthetic procedure with broad applicability. For example, oxidation of cyclopentanone 1a with 2 equiv of Mn(OAc)₃·2H₂O and 1 equiv of Cu(OAc)

Full text of DOI:10.1021/jo961199u

7832
J . Org. Chem. 1996, 61, 7832-7847
Mn (III)-Ba sed Oxid a tive F r ee-Ra d ica l Cycliza tion s of Un sa tu r a ted
Keton es
Bridget McCarthy Cole, Luning Han, and Barry B. Snider*
Department of Chemistry, Brandeis University, Waltham, Massachusetts 02254-9110
Received J une 24, 1996X
Mn(III)-based oxidative free-radical cyclization of unsaturated ketones is a versatile synthetic
procedure with broad applicability. For example, oxidation of cyclopentanone 1a with 2 equiv of
Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O in AcOH at 80 °C for 1.5 h affords 75% of bicyclo-
[3.2.1]oct-3-en-8-one 8a and 15% of bicyclo[3.2.1]oct-2-en-8-one 9a . Bridged bicyclic ketones that
cannot enolize further are isolated in good yield. Monocyclic â,γ-unsaturated ketones that can
enolize are oxidized further to give γ-acetoxy enones. The formation of bicyclo[3.3.1]non-2-en-9-
one (57a ) in 52% yield from 2-allylcyclohexanone (56a ) suggests that kinetically controlled
enolization is the rate-determining step in R-keto radical formation. A wide variety of examples
delineating the scope, limitations, and stereoselectivity of this reaction are presented.
During the past decade, Mn(III)-based oxidative free-
radical 4a , which is oxidized by Cu(II) to give 90% of a
5:1 mixture of 8a and 9a . We have previously observed
that Cu(II) oxidizes alkyl radicals selectively to afford the
least hindered alkene.5 Since the double bonds in both
8a and 9a are disubstituted, oxidative elimination ap-
parently selectively removes the least hindered proton,
which is adjacent to the bridgehead hydrogen rather than
the ester group, to give predominantly 8a . A trace of R,â-
unsaturated enone 6a is obtained from oxidation of
R-keto radical 3a . Similar reactions carried out at 60 °C
for 24 h and 25 °C for 11 d provide only 55% and 23% of
similar mixtures of products, respectively. Oxidation of
1a in ethanol or methanol at reflux provides a small
amount of recovered starting material and polymeric
products; reaction of 1a in benzene gives products in
yields comparable to that in acetic acid, but the reaction
requires 68 h to go to completion. Thus, the optimal
conditions are 80 °C in acetic acid.
Cyclohexanone 1b reacts more slowly (16 h, 80 °C) than
1a to give a similar mixture of 66% of bicyclo[3.3.1]non-
3-en-9-one 8b,6 7% of bicyclo[3.3.1]non-2-en-9-one 9b,7
and 4% of 6-methylenebicyclo[3.2.1]octan-8-one 10b,8
which is formed by 5-exo-cyclization. Heiba and Dessau
have reported that the rate-determining step in R-keto
radical formation from ketones involves reaction of Mn-
(III) with the enol or enolate.4 The rate is independent
of Mn(III) concentration, first order in ketone concentra-
tion, and accelerated by acetate ion, which suggests that
base-catalyzed enolization is the rate-determining step
of the reaction. The observation that cyclohexanone 1b
reacts 10 times slower than cyclopentanone 1a is con-
sistent with this analysis, since enolization of cyclohex-
anone is about 10 times slower than enolization of
cyclopentanone.9
radical cyclizations have been developed into a versatile
protocol for the formation of highly functionalized prod-
ucts from simple precursors.1 These cyclizations have
been initiated by reaction of relatively acidic compounds,
such as 1,3-diketones, acetoacetates, malonates, and
R-sulfinyl or R-nitro ketones, with Mn(OAc)3 to form Mn-
(III) enolates, which undergo electron transfer to give
Mn(II) and radicals. The Mn(III)-based oxidative free-
radical addition of simple symmetrical ketones to simple
alkenes is well known.1c-e In these reactions, the ketone
must be used in large excess to prevent further oxidation
of the product. Accordingly, the reported yields are based
on oxidant consumed, and these reactions are of limited
preparative value. Oxidative free-radical cyclizations of
unsaturated ketones have not been examined, presum-
ably because of anticipated problems with both the
regioselectivity of radical formation and further oxidation
of the cyclic ketone products when the starting ketone is
too valuable to use in large excess.
Cycliza tion of 5-Hexen yl Ra d ica ls. Oxid a tive
Cycliza t ion of 1-Allyl-2-oxocycloa lk a n eca r b oxy-
la tes 1a -d . We have found that Mn(III)-based oxidative
free-radical cyclization of unsaturated ketones is in fact
a versatile synthetic procedure with broad applicability.2
For instance, reaction of cyclopentanone 1a as a 0.1 M
solution in AcOH with 2.25 equiv of Mn(OAc)3‚2H2O and
1 equiv of Cu(OAc)2‚H2O for 1.5 h at 80 °C affords 75%
of bicyclo[3.2.1]oct-3-en-8-one 8a ,3 15% of bicyclo[3.2.1]-
oct-2-en-8-one 9a , and <1% of cyclopentenone 6a . The
slow step appears to be the formation of Mn(III)-enolate
2a ,4a,b which undergoes electron transfer with loss of Mn-
(II) to give R-keto radical 3a . 6-endo-Cyclization affords
Similarly, cycloheptanone 1c (85 h, 80 °C) gives 21%
of bicyclo[4.3.1]dec-7-en-10-one 8c, 1% of bicyclo[4.3.1]-
non-8-en-10-one 9c, and 22% of 7-methylenebicyclo[3.2.1]-
nonan-9-one 10c. As the ring size of the ketone in-
X Abstract published in Advance ACS Abstracts, October 1, 1996.
(1) For reviews see: (a) Snider, B. B. Chem. Rev. 1996, 96, 339. (b)
Iqbal, J .; Bhatia, B.; Nayyar, N. K. Chem. Rev. 1994, 94, 519. (c)
Melikyan, G. G. Synthesis 1993, 833. (d) Badanyan, Sh. O.; Melikyan,
G. G.; Mkrtchyan, D. A. Russ. Chem. Rev. 1989, 58, 286; Uspekhi Khim.
1989, 58, 475. (e) de Klein, W. J . In Organic Synthesis by Oxidation
with Metal Compounds; Mijs, W. J ., de J onge, C. R. H., Eds.; Plenum
Press: New York, 1986; pp 261-314.
(2) For preliminary communications on portions of this work see:
(a) Snider, B. B.; McCarthy Cole, B. J . Org. Chem. 1995, 60, 5376. (b)
Snider, B. B.; McCarthy, B. A. In Benign by Design. Alternative
Chemical Synthesis for Pollution Prevention; Anastas, P., Farris, C.,
Eds.; ACS Symposium Series; American Chemical Society: Washing-
ton, DC, 1994; Chapter 7.
(4) (a) Heiba, E. I.; Dessau, R. M. J . Am. Chem. Soc. 1971, 93, 524.
(b) Fristad, W. E.; Peterson, J . R. J . Org. Chem. 1985, 50, 10. (c) Heiba,
E. I.; Dessau, R. M. J . Am. Chem. Soc. 1972, 94, 2888. (d) Dessau, R.
M.; Heiba, E. I. J . Org. Chem. 1974, 39, 3457.
(5) Snider, B. B.; Kwon, T. J . Org. Chem. 1990, 55, 1965.
(6) Colvin, E. W.; Parker, W. J . Chem. Soc. 1965, 5764.
(7) Kende, A. S; Roth, B.; Sanfilippo, P. J . J . Am. Chem. Soc. 1982,
104, 1784. Kende, A. S.; Wustrow, D. J . Tetrahedron Lett. 1985, 26,
5411.
(3) Buchanan, G. L.; Curran, A. C. W.; McCrae, J . M.; McLay, G.
W. Tetrahedron 1967, 23, 4729.
(8) Berkowitz, W. F.; Wilson, P. J . J . Org. Chem. 1991, 56, 3097.
S0022-3263(96)01199-1 CCC: $12.00 © 1996 American Chemical Society

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7833
E-isomer 1e, indicating that the double bond stereochem-
istry has a remarkable effect on the 6-endo-cyclizations
of R-keto radicals. The E-isomer 1e reacts similarly to
1a , giving 73% of 8e13 and 10% of 9e. Trace amounts of
cyclopentenone 6e and R-acetoxycyclopentanone 7e are
obtained from oxidation of R-keto radical 3e. The ster-
eochemistry of 9e was assigned based on the olefinic
hydrogen absorptions at δ 6.01 (J ) 9.3, 2.6 Hz) and δ
5.45 (J ) 9.3, 2.1, 1.2 Hz). The allylic coupling of 2.6 Hz
and the vicinal coupling of 2.1 Hz indicate that the allylic
methine hydrogen is pseudoaxial.
We were surprised to find that oxidation of 1f with 3
equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O
in AcOH at 80 °C for 2 h only gives 13% of 8e ) 8f along
with 4% of enone 6f and 6% of a 1:1 mixture of acetates
7f. A trace amount of 9f (≈1%) was also detected in the
1H NMR spectrum of the crude mixture of products. This
compound displays an olefinic absorption at δ 6.05 (J )
9.9) and a doublet at δ 1.11 (J ) 7.1), indicating that it
is the stereoisomer of 9e.
The difference in the yield of the bicyclic products from
the Z- and E-isomers 1e and 1f initially led us to suspect
that manganese enolates 2e and 2f cyclize to bicyclic
radicals 4e and 4f, respectively, without the intermediacy
of â-keto radicals 3e and 3f. However, a control experi-
ment established that both the Z- and E-isomers are
oxidized by Mn(III) at the same rate. Ketones 1e and 1f
were simultaneously added to separate, stirred mixtures
of 2 equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu-
(OAc)2‚H2O in acetic acid at 80 °C, and the reactions were
then monitored for the complete consumption of Mn(III)
as indicated by the solution turning bright blue. All the
Mn(III) was consumed in 75 min in both reactions. GC
analysis indicated that 10% of 1e and 33% of 1f re-
mained.
Radicals 3e and 3f are probably intermediates since
the rate of oxidation of 1e and 1f by Mn(III) should be
the same if enolization is the rate-limiting step. The
reaction rates would be very different if cyclization of Mn-
(III) enolate 2e,f to give 4e,f was the rate-determining
step. Since 1e and 1f are oxidized by Mn(III) to 3 at the
same rate, the methyl group of the (Z)-crotyl substrate
must sterically hinder 6-endo-addition of the R-keto
radical of 3f to the double bond, so that 3f gives more 6f,
7f, and polymer.
Oxid a tive Cycliza tion of 1-(2-Bu ten yl)-2-oxocy-
cloh exa n eca r boxyla tes 12a ,b. The oxidative cycliza-
tion of 2(E)-butenyl-substituted ethyl 2-oxocyclohexane-
carboxylate 12a with 2.25 equiv of Mn(OAc)3‚2H2O and
1 equiv of Cu(OAc)2‚H2O in AcOH at 80 °C for 16 h
provides 18% of 16, <1% of 17, 13% of 18, 32% of 20,14
7% of 21a , 7% of 23a , and 2% of 24a . R-Keto radical 13a
undergoes 5-exo-cyclization to provide radicals 14 and 15
and 6-endo-cyclization to afford radical 19a . Oxidation
of 14 and 15 by Cu(II) gives the less hindered terminal
alkenes 16 and 18, respectively, and a trace of 17.
Oxidation of radical 19a by Cu(II) provides a 4:1 mixture
of 20 and 21a , while a 1,5-hydrogen shift gives radical
22a , which is oxidized by Cu(II) to afford 23a and 24a .
The 1,5-hydrogen shift is particularly facile since the
hydrogen is in close proximity to the radical center.
Related examples of cyclooctyl radicals undergoing a
similar 1,5-shift are known.15
creases, 5-exo-cyclization competes more effectively with
6-endo-cyclization. These reactions proceed in syntheti-
cally useful yield since ketones 1a -c can enolize in only
one direction and bridged bicyclic ketones 8a -c, 9a -c,
and 10b ,c are not susceptible to further oxidation
because they cannot enolize.
Cylooctanone 1d behaves very differently (8 equiv of
Mn(OAc)3‚2H2O, 1 equiv of Cu(OAc)2‚H2O, 226 h, 80 °C),
giving 62% of 9-acetoxybicyclo[5.3.1]oct-7-en-11-one 11d .
The larger ring permits enolization10 of the anticipated
major product 8d to give a Mn(III) enolate with an
E-double bond in an eight-membered ring. Oxidation of
the allylic radical to the cation and reaction of the cation
with solvent gives acetate 11d as a single stereoisomer
resulting from attack of acetate on the less hindered exo-
face. The couplings to CHOAc (J ) 6.5, 6.3, 4.8 Hz)
correspond to that calculated for the exo isomer (J ) 8.6,
7.4, 4.2 Hz), but not for the endo isomer (J ) 7.3, 3.2, 1.2
Hz).11 The facile oxidation of â,γ-unsaturated enones to
γ-acetoxy-R,â-unsaturated enones by Mn(III) and Cu(II)
has been extensively studied.12
Oxid a tive Cycliza tion of Eth yl 1-(2-Bu ten yl)-2-
oxocyclop en ta n eca r boxyla tes (1e,f). 2(Z)-Butenyl-
cyclopentanone 1f behaves very differently than the
(9) Cyclopentanone is deprotonated by triethylamine in DMF/water
10 times faster than cyclohexanone: (a) Toullec, J . Enolisation of
Simple Carbonyl Compounds and Related Reactions. In Advances in
Physical Organic Chemistry; Gold V., Ed.; Academic Press: London,
1982; Vol. 18, p 1. (b) Shechter, H.; Collis, M. J .; Dessy, R.; Okuzumi,
Y.; Chen, A. J . Am. Chem. Soc. 1962, 84, 2905.
(10) Buchanan, G. L.; J amieson, G. Tetrahedron 1972, 28, 1123 and
1129.
(11) Model version KS 2.99 obtained from Prof. Steliou, Boston
University, was used for MM2 calculations.
(13) Evans, E. H.; Hewson, A. T.; March, L. A.; Nowell, I. W.;
Wadsworth, A. H. J . Chem. Soc., Perkin Trans. 1 1987, 137.
(14) (a) Erman, W. F.; Kretschmar, H. C. J . Org. Chem. 1968, 33,
1545. (b) Foote, C. S.; Woodward, R. B. Tetrahedron 1964, 20, 687.
(12) Breuilles, P.; Uguen, D. Bull. Soc. Chim. Fr. 1988, 88, 705.

7834 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
radical 15 with an endo-methyl group. Oxidative elimi-
nation of cyclohexyl radicals 19a and 22a selectively
removes the least hindered proton, which is adjacent to
the bridgehead hydrogen rather than the ester group, to
give predominantly 20 and 23a , as we observed in the
oxidation of 4b.
The structures of the products were assigned by
analysis of the 1H NMR spectral data. The allylic
methine hydrogen of 16 at δ 2.66 (J ) 4.6, 8.3, 9.1 Hz) is
coupled to the vinyl hydrogen with J ) 8.3 Hz and to
the adjacent methylene group with J ) 4.6 and 9.1 Hz.
Therefore, the coupling to the bridgehead hydrogen is less
than 1 Hz, which is consistent with that calculated for
the endo allylic hydrogen of 16 (dihedral angle calcu-
lated11 to be 94°) but not for the exo allylic hydrogen of
18 (dihedral angle calculated11 to be 35°). The structures
of products 21a ,b, 23a ,b, and 24a ,b were assigned by
analogy to 8b and 9b.16 The stereochemistry of 21a was
assigned analogously to that of 9e; the allylic coupling
of 2.3 Hz and the vicinal coupling of 2.7 Hz indicate that
the allylic methine hydrogen is pseudoaxial.
Oxid a tive Cycliza tion of 1-P r en yl-2-oxocycloa l-
k a n eca r boxyla tes 25a ,b. Treatment of 25a with 3
equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O
in AcOH at 80 °C for 2.5 h provides only 9% of bicyclo-
[3.2.1]octenone 26a along with 2% of bicyclo[2.2.1]-
heptanone 28a and 6% of cyclopentenone 27a . R-Keto-
cyclopentyl radicals such as 3a and 3e undergo 6-endo-
cyclization if there is no Z-substituent on the double bond.
If there is a Z-substituent on the double bond as in 3f,
6-endo-cyclization is retarded and 5-exo-cyclization leads
to strained products. Therefore, bicyclic products are
obtained in low yield and oxidation to enone 27a and
polymerization are the major processes.
The 5-exo-cyclization of R-ketocyclohexyl radicals is
more facile, so that 5-exo-cyclization of radical 13b
proceeds in good yield since 6-endo-cyclization is retarded
by the cis substituent on the alkene. As expected,
oxidative cyclization of 1-prenyl-2-oxocyclohexanecar-
boxylate 25b with 2.5 equiv of Mn(OAc)3‚2H2O and 1
equiv of Cu(OAc)2‚H2O in AcOH at 80 °C for 18 h provides
51% of 28b, 1% of 29b, and 1% of 26b. The stereochem-
istry of 28b was assigned on the basis of the ≈0 Hz
coupling constant between the allylic methine at δ 2.67
and the bridgehead hydrogen, which is consistent with
that expected for the allylic endo hydrogen.
More of the rearrangement products 23a and 24a
should be formed if less Cu(II) is used, since the oxidation
of 19a is a second-order process and the rearrangement
of 19a to 22a is a first-order process. As anticipated,
oxidation of 12a with 2.25 equiv of Mn(OAc)3‚2H2O and
only 0.1 equiv of Cu(OAc)2‚H2O provides 15% of 16, 10%
of 18, 26% of 20,14 6% of 21a , 18% of 23a , and 5% of 24a .
As expected, the yields of rearrangement products 23a
and 24a increase significantly, at the expense of 20 and
21a , which confirms that they are obtained from oxida-
tion of 22a , which is formed from radical 19a by a 1,5-
hydrogen shift.
Oxidation of Z-substrate 12b with 2.25 equiv of Mn-
(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O in AcOH at
80 °C for 16 h provides 46% of 16, <1% of 17, 5% of 18,
2% of 20, 2% of 21b, 2% of 23b, and 1% of 24b. As
observed in the oxidative cyclization of 1f, the Z-methyl
group of 12b obstructs 6-endo-addition of the radical to
the double bond so that 5-exo-cyclization becomes the
major reaction pathway, providing bicyclo[3.2.1]octanone
16 as the major product in synthetically useful yield. The
methyl group stereochemistry also affects the stereo-
chemistry of the 5-exo-cyclization. E-Isomer 12a provides
a 1.4:1 mixture of 16 and 18, while the Z-isomer 12b
provides a 9:1 mixture of 16 and 18 indicating that the
Z-methyl group hinders the 5-exo-cyclization that gives
Oxidative Cyclization of 1-Meth allyl-2-oxocycloal-
k a n eca r boxyla tes 30a ,b. Oxidative cyclization of 30a
(15) For examples of 1,5-hydrogen shifts in cyclooctyl systems see:
(a) Meek, J . S.; Fowler, J . S. J . Am. Chem. Soc. 1967, 89, 1967. (b)
Matsumoto, H.; Nakano, T.; Takasu, K.; Nagai, Y. J . Org. Chem. 1978,
43, 1734.
(16) For further discussion of structure determination of the cy-
clization products see: Cole, B. M. Ph.D. Thesis, Brandeis University,
1995. Han, L. Ph.D. Thesis, Brandeis University, 1996.

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7835
gives complex product mixtures typical of reactions that
yield tertiary radicals, which are oxidized to cations by
either Mn(III) or Cu(II). Treatment of 30a with 2.25
equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O
in AcOH for 1.5 h at 80 °C provides 6% of 33a , 22% of
34a , 22% of 35a , 5% of acetate 36a , 5% of acetate 37a ,
5% of lactone 42a , and 3% of lactone 44a . Bicyclic radical
31a is oxidized by Mn(III) or Cu(II) to tertiary cation 32a .
Loss of a proton affords alkenes 33a , 34a , and 35a
without the regiocontrol typical of Cu(II) oxidation of
secondary radicals. Addition of acetate to cation 32a
provides acetates 36a and 37a . Since bicyclic radical 31a
is tertiary, it can be oxidized to cation 32a by both Mn-
(III) and Cu(II). Accordingly, similar results are obtained
without Cu(OAc)2‚H2O.
acetate of 37a and the methyl group of 36a are shielded
by the ketone. The acetate of 37a absorbs upfield at δ
1.93 as compared to 36a (δ 2.07), while the methyl group
of 36a absorbs upfield at δ 1.59 as compared to 37a (δ
1.69).
The structures of the unexpected products 42a and 44a
1
were assigned on the basis of their H and 13C NMR and
IR spectral data. Lactone 42a displays signals at δ 179.4
and δ 167.7 in the 13C NMR spectra and absorptions at
1776 and 1710 cm-1 in the IR spectra, indicating the
presence of a γ-lactone and an R,â-unsaturated ester, but
not a cyclopentanone. Although only partial 13C NMR
data were obtained for 44a , we were able to establish
that it contains a γ-lactone and saturated ester on the
basis of the absorptions at 1769 and 1737 cm-1 in the IR
spectra. A plausible mechanism involves addition of the
carbonyl oxygen to the cation of 32a to give cation 38a ,
which is attacked by adventitious water to provide lactol
39a . Oxidative fragmentation of Mn(III) alkoxide 39a ,
as we have previously observed for cyclobutanols17 and
strained cyclopentanols,18 gives radicals 41a and 43a ,
which are oxidized by Cu(II) to give lactones 42a and 44a .
The regioselectivity of the oxidative elimination leading
to 42a and 44a (loss of the more hindered proton) is
opposite to what we have usually observed (loss of the
least hindered proton). It is possible that Cu(II) coordi-
nates to the lactone or that the electron-withdrawing
effect of the lactone makes the adjacent proton more
acidic.
Oxidative cyclization of cyclohexanone 30b with 2.25
equiv of Mn(OAc)3 and 1 equiv of Cu(OAc)2 in AcOH at
80 °C for 18 h affords 27% of 33b,19 32% of 34b,19 13% of
35b, and 2% of the monocyclic R′-acetoxy ketone. Ap-
parently, loss of a proton to give 33b-35b is the exclusive
reaction of bicyclo[3.3.1]nonyl cation 32b, in contrast to
bicyclo[3.2.1]octyl cation 32a , which loses a proton to give
33a -35a , reacts with acetate to give 36a and 37a , or
cyclizes leading eventually to lactones 42a and 44a .
Molecular mechanics calculations indicate that bicyclo-
[3.3.1]nonyl acetates 36b and 37b are relatively more
strained (compared to bicyclo[3.3.1]nonenes 33b-35b)
than bicyclo[3.2.1]octyl acetates 36a and 37a (compared
to bicyclo[3.3.1]octenes 33a -35a ), so that loss of a proton
from 32b is the exclusive reaction. If 39b were formed,
oxidative ring cleavage would not be expected because
cyclohexanols do not undergo Mn(III)-based oxidative
fragmentations.18
Oxid a tive Cycliza tion of 5-Meth ylcyclop en ta n on e
45. All of the examples discussed above involve forma-
tion of the Mn(III) enolate by loss of a proton from a
methylene group. Oxidative cyclization of 45 was exam-
ined in order to determine if substrates with a methine
hydrogen are usable. Treatment of 45 with 3 equiv of
Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O in acetic
acid at 80 °C for 20 h provides 41% of 46, 9% of 47, 3%
of acetate 48, and 24% of a 2:3 mixture of lactones 50
and 51. Although 45 was employed as a 1:1 mixture of
diastereomers, each isomer is oxidized to the same
tertiary R-keto radical, which undergoes 6-endo-cycliza-
tion exclusively. Oxidation of the resulting bicyclic
radical by Cu(II) provides a 4:1 mixture of bicyclo[3.2.1]-
octanones 46 and 47. The structures of 46 and 47 were
The stereochemistry of acetates 36a and 37a was
assigned on the basis of the chemical shift of the methyl
groups. MM2 calculations11 clearly show that both
stereoisomers exist primarily in a boat conformer as
shown. The calculated distance between the ketone and
the exo-substituent is approximately 2.5 Å; therefore, the
(17) Snider, B. B; Vo, N. H.; Foxman, B. M. J . Org. Chem. 1993, 58,
7228
(18) Snider, B. B.; O’Neil, S. V. Unpublished results.
(19) Kende, A. S.; Schneider, J . A. Synth. Commun. 1979, 9, 419.

7836 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
1
established by comparison of their H NMR spectral data
hexanone (56a ) with 4 equiv of Mn(OAc)3‚2H2O and 1
equiv of Cu(OAc)2‚H2O in a 0.5 M solution of KOAc in
acetic acid at 80 °C for 24 h provides bicyclo[3.3.1]non-
2-en-9-one (57a )14 (51%), enone 58a (1%), and R-acetoxy
ketone 59a (3%), which are all formed by enolization to
the kinetically favored methylene side of ketone 56a .
Enolization of the more hindered methine hydrogen
affords a tertiary radical that is oxidized to give enone
61a 20 (1%) and R-acetoxy ketone 60a (3%). Further
oxidation of either acetoxy ketone 59a or 60a leads to
bicyclic acetoxy ketone 62a (4%). The reaction proceeds
similarly in the absence of potassium acetate, but is less
reproducible.
to those of 8a and 8b. As expected, enolization of the
methine hydrogen of 45 is slower than loss of one of the
methylene hydrogens of 1a so that the reaction requires
20 h rather than 1.5 h. Oxidative elimination by Cu(II)
shows modest preference for abstraction of a proton from
the methyl side rather than the ester side of the ring,
possibly indicating electronic as well as steric control of
the Cu(II) oxidative elimination of secondary radicals to
alkenes.
Lactones 49 and 50 were isolated as an inseparable
mixture. The IR spectrum shows two carbonyl stretch-
ings at 1771 and 1731 cm-1, indicating the presence of a
γ-lactone and an ester group. These lactones are ob-
tained from a Baeyer-Villiger-type oxidation of 46.
Since a tertiary allylic group will migrate more readily
than a secondary allylic group, 46 undergoes Baeyer-
Villiger oxidation to give 49 while the analogue without
the methyl group, 8a , is stable. Allylic lactone 49
equilibrates with 50 in AcOH at 80 °C by either a [3,3]-
sigmatropic rearrangement or ring opening and closure.
Oxid a tive cycliza tion of 2-a llyl-2-d od ecylcyclo-
p en ta n on e (51) with 3 equiv of Mn(OAc)3‚2H2O and 1
equiv of Cu(OAc)2‚H2O in acetic acid at 80 °C for 12 h
gives bicyclo[3.2.1]octenones 52 (44%) and 53 (7%),
R-acetoxy ketone 54 (10%), and R-hydroxy ketone 55 (7%),
indicating that the ester group is not needed for the
success of this sequence. The reaction is about 10 times
slower than the oxidative cyclization of 1a , suggesting
that the R-ester group makes the R′-proton of 1a more
acidic than that of 51.
Similarly, oxidative cyclization of 2-allylcycloheptanone
(56b) with 4 equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu-
(OAc)2‚H2O in acetic acid at 80 °C for 60 h affords 32%
of bicyclo[4.3.1]dec-8-en-10-one (57b), 11% of R-acetoxy
ketone 60b, and 10% of recovered 56b. Cycloheptanone
56b reacts slower than cyclohexanone 56a and undergoes
enolization less selectively at the methylene carbon.
Oxidative cyclization of 2-allylcyclooctanone (56c) with
6 equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu(OAc)2‚H2O
in acetic acid at 80 °C for 42 h affords 25% (36% based
on recovered 56c) of 9-exo-acetoxybicyclo[5.3.1]undec-10-
en-11-one (63c). As in the cyclization of 1d to 11d , the
initially formed bicyclic product 57c enolizes,10 and the
resulting dienolate is oxidized to an allylic radical that
is oxidized to acetate 63c.
Oxid a tive Cycliza tion of Bicyclic Keton es. Oxida-
tive cyclization of bicyclic ketones 64 and 67 leads
efficiently to tricyclic ketones 65 and 68. Tetralone 64
cyclizes as expected to give a 14:1 mixture of 65 and 66
in 75% yield with selective formation of the less hindered
alkene. Oxidative cyclization of 67 proceeds in high yield,
affording 80% of 68, 7.5% of 69, and 2.5% of 70. The
radical obtained by enolization and oxidation of octalone
67 could cyclize to either allyl group. The ring fusion
hydrogen, Hc, of the major product 68 absorbs at δ 2.92,
J Ha,Hc ) 0 Hz, J Hb,Hc ) 9.5 Hz, J )CH,Hc ) 5.4 Hz. These
values correspond very closely to those calculated11 for
68 (J ) 1.0, 8.8, 4.6 Hz) and are quite different from those
calculated for the diastereomer (J ) 5.1, 4.5, 2.5 Hz).
Apparently, the R-keto radical cyclizes exclusively to the
R-allyl group with the cyclohexanone in the boat confor-
Oxid a t ive Cycliza t ion of 2-Allylcycloa lk a n on es
56a -d . We believed that 2-allylcycloalkanones 56a -c
should be suitable substrates since enolization, the rate-
determining step, should occur selectively at the less
substituted carbon. Oxidative cyclization of 2-allylcyclo-
(20) Taber, D. F. J . Org. Chem. 1976, 41, 2649.

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7837
mation. Cyclization through the chair conformation
would occur on the â-allyl group to give the less stable
diastereomer of 68. This is not observed due to severe
steric interactions between the methyl group and the
three-carbon bridge in the cyclization transition state.
75b and 4.8 and 6.9 Hz for 74b, as discussed above.
These results indicate that oxidative cyclization of ke-
tones may be useful even when enolization and subse-
quent oxidation of the initial product occur since γ-acetoxy-
R,â-unsaturated ketones are obtained in good yield.
Similar results are obtained with 2-acetyl-2-allylbuty-
rolactone (76a ), which affords 39% of 78a and 24% of 77a ,
and with 2-acetyl-2-crotylbutyrolactone (76b), which
affords 27% of 78b, 17% of 77b, and 2% of 79b. The
stereochemical assignments of 77 and 78 follow from the
coupling constants of the CHOAc proton to the adjacent
methylene group, which are 5.3 and 10.9 Hz for 78a , 5.2
and 10.8 Hz for 78b, and 5.4 and 8.2 Hz for 77a .
Oxida tive Cycliza tion of r-Disu bstitu ted Acetoa c-
eta tes 71a ,b a n d Acetylbu tyr ola cton es 76a ,b. The
initial products of the oxidative cyclization of 71a ,b are
3-cyclohexenones 72a ,b that are rapidly oxidized by Mn-
(III) to provide γ-acetoxy-R,â-unsaturated ketones. Treat-
ment of 71a with 6 equiv of Mn(OAc)3‚2H2O and 1 equiv
of Cu(OAc)2‚H2O in AcOH for 48 h at 80 °C gives 60% of
a 1:1 mixture of 74a and 75a . Oxidative cyclization of
methyl ketone 71a should give â,γ-unsaturated ketone
72a , which is not isolated. 3-Cyclohexenone 72a is
oxidized rapidly to radical 73a because â,γ-unsaturated
ketone 72a is more acidic than 71a . Oxidation of radical
73a provides γ-acetoxy-R,â-unsaturated ketones 74a and
75a as the only isolable products. Oxidation of â,γ-
unsaturated ketones by Mn(III) to γ-acetoxy-R,â-unsatur-
ated ketones is well known.12
The stereochemistry of 74a and 75a was assigned on
the basis of the coupling constants of the CHOAc proton
to the adjacent methylene group, which are 5.4 and 10.0
Hz for 75a and 5.1 and 7.8 Hz for 74a . MM2 calcula-
tions11 suggest that this proton should be exclusively
axial, with larger coupling constants, in 75a , in which
the acetate and the larger, methyl group are cis. In 74a ,
with the acetate and smaller, ester group cis, both
conformers are populated so that this proton is equatorial
a significant percentage of the time.21
Oxid a tive Cycliza tion of 1-P h en yl-6-n on en -1-on e
(80). Several years ago we reported that oxidative
cyclization of silyl enol ether 82 with either Ce(IV) or Cu-
(II) provides tricyclic ketones 81a ,b in excellent yield.22
These reactions proceed via cation radicals rather than
radicals and provide 81a ,b in excellent yield because the
silyl enol ether 82 is more easily oxidized than the
product ketone 81a ,b. We examined the oxidative cy-
clization of ketone 80 to determine whether the expected
initial product, tricyclic ketone 81a ,b, is stable to the
conditions required for Mn(III) oxidation of 80. Oxidative
cyclization of 80 with 6 equiv of Mn(OAc)3‚2H2O and 1
equiv of Cu(OAc)2‚H2O in AcOH for 60 h at 80 °C provides
15% of 83a , 9% of 83b, 9% of 88, and none of 81a ,b.
Apparently, oxidation of 81a ,b to radical 84 is faster than
oxidation of 80. Further oxidation of radical 84 will
provide acetates 83a ,b and phenol 85, which is oxidized
to acetoxy enone 88.23 The structures of 83a ,b were
assigned on the basis of the similarity of the proton and
carbon NMR spectra to those of 81a ,b.
Oxidative cyclization of R-crotylacetoacetate 71b pro-
ceeds similarly, affording 47% of a 2:1 mixture of 74b
and 75b. The stereochemistry was assigned on the basis
of the coupling constants of the CHOAc proton to the
adjacent methylene group, which are 5.4 and 9.6 Hz for
We established that tricyclic ketones 81a ,b are com-
petent intermediates by treating 81a ,b22 with 2 equiv of
(21) The observed coupling constants are similar to those observed
for related 4-acetoxy-2-cyclohexenones: Widmer, E.; Zell, R.; Lukac,
T.; Casadei, M.; Schonholzer, P.; Broger, E. A. Helv. Chim. Acta 1981,
64, 2405.
(22) Snider, B. B.; Kwon, T. J . Org. Chem. 1992, 57, 2399.
(23) For related oxidations of phenols see: Greenland, H.; Pinhey,
J . T.; Sternhell, S. Aust. J . Chem. 1986, 39, 2067; 1987, 40, 325.

7838 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
R-keto radicals by either Cu(II) or Mn(III) is inefficient17
so that cyclohexenone 96 is obtained in low yield.
Cycliza tion of 6-Hep ten yl a n d 7-Octen yl Ra d ica ls.
Oxid a tive Cycliza tion s of 97a -c. The oxidative cy-
clizations of 97a -c, 102a -c, 105a ,b, and 111 were
carried out to establish whether medium-sized rings
could be obtained from oxidative cyclization of ketones.
Treatment of 97a with 2.5 equiv of Mn(OAc)3‚2H2O and
1 equiv of Cu(OAc)2‚H2O in AcOH for 2.5 h at 80 °C
provides 7-endo-cyclization products 98a (25%) and 99a
(10%) and 6-exo-cyclization product 100a (25%) and
lactone 101a (12%), which results from a Baeyer-
Villiger-type oxidation of 100a . Thus, the monocyclic
R-keto radical obtained from 97a undergoes 6-exo- and
7-endo-cyclization in approximately equal amounts.
We were surprised to isolate lactone 101a , which is
much more polar than ketone 100a . Ketone 100a
displays carbonyl signals at δ 210.1 and 171.2 in the 13C
NMR spectrum and absorptions at 1742 and 1712 cm-1
in the IR spectrum, whereas lactone 101a exhibits signals
at δ 171.6 and 170.9 in the 13C NMR spectrum and
absorptions at 1770 and 1732 cm-1 in the IR spectrum,
indicating the presence of a γ-lactone in addition to the
ethyl ester. We determined that lactone 101a is formed
by a Baeyer-Villiger oxidation by treating ketone 100a
with 2 equiv of Mn(OAc)3‚2H2O in AcOH for 24 h to
provide a 1:1 mixture of 100a and 101a .
At first we were puzzled as to why 100a is oxidized to
a lactone while 98a , 99a , and all previously reported
ketones are not, except for 46, which has a tertiary alkyl
group that migrates more readily. An allyl group mi-
grates readily in a Baeyer-Villiger oxidation if the
torsion angle between the double bond and the migrating
bond is ≈90°, so that the double bond stabilizes the
migrating substituent. Examination of models suggests
that Baeyer-Villiger reaction of 100a occurs readily
because the torsion angle H2CdCCHC in the stable
conformer with a chair cyclohexanone is 118°, which is
close to optimal. Baeyer-Villiger reaction of endocyclic
analogues 8 and 98a should be slower since the torsion
angle CH)CCHC is 29° and 6°, respectively, which is far
from the optimal 90° for migration of an allylic group.
Baeyer-Villiger reaction of exocyclic homologue 100b is
slow since the cyclohexanone is calculated to be more
stable in the boat conformer with a H2CdCCHC torsion
angle of 21°. Other examples of Baeyer-Villiger oxida-
tion of bridged bicyclic ketones are known.27
Mn(OAc)3 and 1 equiv of Cu(OAc)2 in AcOH at 80 °C for
48 h to afford 11% of 83a , 9% of 83b, 13% of 88, and
31% of a 3:2 mixture of 86 and 87. R-Ketol 86 could be
formed by oxidation of radical 84 and reaction with water
rather than acetate or by hydrolysis of 83b. Hydrolysis
of 83b with NaOH in EtOH affords 82% of a 1:1 mixture
of 86 and 87. The bridged ketol 87 is formed from the
fused ketol 86 by the R-ketol rearrangement.24
Oxid a t ive Cycliza t ion of 2-Allylic-1,3-cycloa l-
k a n ed ion es. Oxidative cyclization of 2-allyl-2-methyl-
cyclopentane-1,3-dione (89a ) in AcOH at 80 °C for 2 h
provides 75% of 90a , 9% of 91a , and 2% of enedione 92a .
Similarly, 2-crotyl-2-methylcyclopentane-1,3-dione (89b)
provides 33% of 90b,25 23% of 91b, and 6% of enedione
92b. Although bicyclo[3.2.1]octenediones 90 and 91 could
enolize and be oxidized further, they are apparently less
acidic than the starting dione 89 and do not react further.
Thus, while 71 gives only acetoxy enones 74 and 75
resulting from further oxidation of the initial product 72,
in some cases it is possible to isolate products that can
enolize. Oxidative cyclization of 89a at 80 °C provides a
much higher yield of 90a and 90b than oxidative cycliza-
tion of 4-allyl-2-methylcyclohexanone at 25 °C for 24 h,
which gives the same bicyclic radical, but provides only
38% of 90a .26
Oxidative cyclization of cyclohexanone 97b affords
7-endo-cyclization products 98b (6%) and 99b (37%)26 and
6-exo-cyclization product 100b (40%). We were surprised
to discover that the Cu(II)-mediated oxidative elimination
of the cycloheptyl radicals obtained from 97a and 97b
proceeds with opposite regioselectivity. Cu(II) oxidation
affords â,γ-unsaturated cycloheptanone 98a as the major
product from cyclopentanone 97a but γ,δ-unsaturated
Oxidative cyclization of cyclohexanedione 93 affords
17% of conjugated cyclohexenone 96 as the only isolable
product. Apparently, the initially formed bicyclic radical
94 undergoes a facile transannular 1,5-hydrogen shift15
to give the more stable R-keto radical 95. Oxidation of
(24) For related R-ketol rearrangements see: Grunewald, G. L.;
Walters, D. E.; Kroboth, T. R. J . Org. Chem. 1978, 43, 3478.
(25) Kasturi, T. R.; Reddy, S. M. Ind. J . Chem. 1982, 21b, 901.
(26) Kates, S. A.; Dombroski, M. A.; Snider, B. B. J . Org. Chem.
1990, 55, 2427.
(27) Krow, G. R. Tetrahedron 1981, 37, 2697. Mehta, G.; Pandey,
P. N.; Ho, T.-L. J . Org. Chem. 1976, 41, 953. Miura, H.; Hirao, K.-I.;
Yonemitsu, O. Tetrahedron 1978, 34, 1805.

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7839
cycloheptanone 99b as the major product from cyclohex-
anone 97b.
We were disappointed to find that oxidative cyclization
of bishomoallylic cyclopentanone 97c affords only 1% of
98c, 10% of 99c, and 2% of 100c. Neither 8-endo- nor
7-exo-cyclization is an efficient process.
further oxidation of 112. γ,δ-Unsaturated cyclohexanone
112 is oxidized at about the same rate as the starting
ketone 111. The stereochemistry of 112 was assigned
based on the chemical shift of the methyl group at δ 1.27,
which corresponds closely to that reported for an equato-
rial methyl group (δ 1.28) but not an axial methyl group
(δ 1.45) in an analogous system.28
Oxidative cyclization of 3(Z)-hexenylcyclopentanone
102a affords 72% of a 5:2 mixture of 6-exo-cyclization
products 103a and 104a . Similarly, cyclohexanone 102b
provides 66% (75% based on recovered 102b) of a similar
mixture of 103b and 104b. The stereochemistry of 103a
follows from the coupling of the allylic methine Ha at δ
2.70 to the adjacent methylene group with J ) 10.6 and
6.3 Hz, which establishes that the hydrogen is axial and
the propenyl side chain equatorial. The spectral data of
the minor isomer 104a are very similar except for the
changes expected for a Z-double bond. Cyclization to the
1,2-disubstituted double bond occurs exclusively 6-exo
through a chair transition state with an equatorial
alkenyl group. Surprisingly, Cu(II) oxidation affords a
5:2 E/ Z mixture of double bonds, while closely related
examples with an ester at the bridgehead give exclusively
the E-double bond.26 We therefore examined the cycliza-
tion of the R′-methyl analogue 102c, which provides 72%
of an 8:1 mixture of 103c and 104c with much better
selectively for the E-isomer, indicating that the stereo-
chemistry of the oxidative elimination of radicals to
alkenes by Cu(OAc)2 is very sensitive to the local steric
environment.
Oxidative cyclization of benzyl cyclopentanone 114
affords 35% of 115, indicating that R-keto radicals
generated by this procedure cyclize onto aromatic rings.
Oxidation of the R-keto radical formed from 114 gives
10% of the enone analogous to 6 and 39% of the R-ac-
etoxycyclopentanone analogous to 7, indicating that, as
expected,29 cyclization of the phenylbutyl radical obtained
from 115 is much slower than cyclization of 5-hexenyl
radical 3. Attempted oxidative cyclization of â-keto ester
116 gives none of the desired product 117. This suggests
that R-keto radicals are more reactive than radicals
obtained from â-dicarbonyl compounds, which are stabi-
lized by two adjacent carbonyl groups.
Ta n d em Cycliza tion s. The selective conversion of
acetoacetate 118 to either propenylcyclohexanone 119 or
vinylnorbornanone 123 at different temperatures clearly
illustrates the difference in reactivity between 1,3-
dicarbonyl compounds and ketones. We have previously
reported the oxidative cyclization of 118 at 25 °C to give
119 (56%), 120 (3%), and 121 (14%); no 123 is obtained
at room temperature.1a,26 On the other hand, oxidation
of 118 with 4 equiv of Mn(OAc)3‚2H2O and 1 equiv of Cu-
(OAc)2‚H2O for 56 h at 60 °C or 16 h at 80 °C provides
45% of a 1:9 mixture of 122 and 123, only 10% of 119,
4% of a 1:2 mixture of 120 and 121, and a trace of bicycle
Oxidative cyclization of acetoacetate 105a yields 7-endo-
cyclization product 107a (1%) and 6-exo-cyclization prod-
ucts 109a (32%) and 110a (10%), which are both derived
from further oxidation of â,γ-unsaturated ketone 108a .
Thus, 6-exo-cyclization of the acyclic R-keto radical is the
predominant process. Oxidative cyclization of 105b
provides 8-endo-cyclization products 106b (8%) and 107b
(19%) and 7-exo-cyclization product 109b (4%).
(28) Sugai, T.; Kakeya, H.; Ohta, H.; Morooka, M.; Ohba, S.
Tetrahedron 1989, 19, 6135.
(29) Reaction of the 4-phenylbutyl radical with 0.025
M Cu-
(OAc)2‚H2O gives 10% of cyclized product and 86% of oxidized
product.30a Therefore, 0.025(kox/kr1) ) 86/10 ) 8.6 or kox/kr1 ) 8.6/0.025
) 344. Reaction of the 5-hexenyl radical with 0.02 M Cu(OAc)2‚H2O
gives 84% of cyclized product and 9.5% of oxidized product.30b There-
fore, 0.02 × kox/kr2 ) 9.5/84 ) 0.113 or kox/kr2 ) 0.113/0.02 ) 5.65.
Therefore, kr2/kr1 ) 344/5.65 ) 60.8 so that cyclization of a 5-hexenyl
radical is about 60 times faster than cyclization of a 4-phenylbutyl
radical.
Oxidative cyclization of 111 proceeds stereospecifically,
but in modest yield, giving 14% of cyclohexanone 112 and
10% of acetoxycyclohexanone 113, which is formed by
(30) (a) Kochi, J . K.; Gilliom, R. D. J . Am. Chem. Soc. 1964, 86, 5251.
(b) J enkins, C. L.; Kochi, J . K. J . Am. Chem. Soc. 1972, 94, 843.

7840 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
reospecifically as previously discussed in the cyclization
of 102c. A second, 5-exo-cyclization followed by oxidation
of the resulting primary radical affords 133 and 134. The
stereochemistry of the ethyl group is tentatively assigned
on the basis of MM2 calculations11 that suggest 133 is
less strained.
124 derived from oxidative cyclization of 122. Oxidation
of pure 119 under these conditions provides 75% of 123
and 8% of enone 122, while oxidation of the 1:2 mixture
of 120 and 121 with 2 equiv of Mn(OAc)3‚2H2O and 1
equiv of Cu(OAc)2‚H2O in AcOH at 80 °C for 16 h affords
70% of a 1:2 mixture of 124 and 125.
At elevated temperatures cyclohexanone 119 is oxi-
dized to the R-keto radical, which undergoes a 5-exo-
cyclization stereospecifically to give the less hindered
secondary radical, which is oxidized by Cu(II) to afford
123 with the less substituted double bond.5 Decoupling
established that the allylic hydrogen of 123 at δ 2.88 was
weakly coupled to the endo hydrogens at δ 1.70 and 1.48,
but not the exo hydrogens at δ 2.06 and 1.76. A long-
range coupling between the allylic methine and the endo
hydrogens requires a W relationship, which is only
possible for the anti isomer 123.
Con clu sion . These results vastly extend the scope of
Mn(III)-based oxidative cyclizations beyond 1,3-dicarbon-
yl compounds. A wide variety of unsaturated ketones can
now be used as substrates. The formation of 57a sug-
gests that kinetically controlled enolization is the rate-
determining step in R-keto radical formation. Bridged
bicyclic ketones that cannot enolize further are isolated
in good yield. â,γ-Unsaturated monocyclic ketones that
can enolize are oxidized to γ-acetoxy enones.
Exp er im en ta l Section
Gen er a l P r oced u r es. 1H NMR spectra were recorded at
300 MHz in CDCl3. Chemical shifts are reported in δ and
coupling constants in Hz. IR spectra are reported in cm-1
.
All air-sensitive reactions were run under N2 in flame-dried
glassware with magnetic stirring.
Oxid a tive Cycliza tion of Eth yl 2-Oxo-1-(2-p r op en yl)-
cyclop en ta n eca r boxyla te (1a ). A solution of â-keto ester
1a (98 mg, 0.5 mmol), Mn(OAc)3‚2H2O (303 mg, 1.125 mmol),
and Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed
AcOH was stirred under N2 for 1.5 h at 80 °C. The mixture
was diluted with water (20 mL), and a 10% solution of NaHSO3
was added dropwise until all residual Mn(III) was reduced to
Mn(II). The resulting solution was extracted with three 20
mL portions of CH2Cl2. The combined extracts were washed
with saturated NaHCO3 and brine solutions. The solvent was
removed under reduced pressure to yield 96 mg of a pale yellow
oil. Purification by flash chromatography (20:1 hexane/EtOAc)
gave 74 mg of an 8:1 mixture of 8a and 9a , followed by 15 mg
of a 10:5:1 mixture of 8a , 9a , and 6a and 6 mg of polymeric
material. The calculated yields are e1% of 6a , 75% of 8a , and
15% of 9a .
Attempted tandem cyclization of 126 was unsuccessful,
giving only the bicyclic products 128 (23%) and 129 (5%),
which are formed analogously to 8e and 9e, and none of
the desired tricyclic product 130. Apparently, oxidation
of bicyclic radical 127 is faster than the second cycliza-
tion. No 130 was obtained even when the reaction was
repeated with 2 equiv of Mn(OAc)3‚2H2O and only a
catalytic amount (5 mol%) of Cu(OAc)2‚H2O to slow down
the oxidation of radical 127. This suggests that cycliza-
tion of bicyclic radical 127, which would have to occur
on the sterically hindered endo face, is very slow.
Tandem oxidative cyclization of 131 affords 82% of a
9:1 mixture of tricyclic ketones 133 and 134. Oxidation
and 6-exo-cyclization affords bicyclic radical 132 ste-
1
A trace of enone 6a was observed in the H NMR spectrum
of the mixture: 7.75 (ddd, 1, J ) 2.4, 2.4, 5.7), 6.16 (br d, 1, J
) 5.7).

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7841
The data for 8a : 1H NMR 5.84 (dddd, 1, J ) 9.3, 6.7, 2.5,
0.8), 5.62 (ddd, 1, J ) 9.3, 4.0, 2.5), 4.23 (q, 2, J ) 7.2), 3.30
(dddd, 1, J ) 17.7, 2.5, 2.5, 2.5), 2.69 (br dd, 1, J ) 17.7, 4.0),
2.69-2.62 (m, 2), 2.14-2.09 (m, 3), 1.29 (t, 3, J ) 7.2); 13C
NMR 211.1, 171.2, 131.7, 125.5, 61.3, 55.8, 46.3, 44.7, 30.9,
28.9, 14.2; IR (neat) 1759, 1728. The data are identical to those
previous reported.3
EtOAc/CH2Cl2) gave 22 mg (10%) of unreacted 1d , followed
by 172 mg (62%) of 11d : 1H NMR 6.32 (d, 1, J ) 4.8), 5.56
(dddd, 1, J ) 6.5, 6.3, 4.8, 1.0), 4.24 (q, 2, J ) 7.1), 2.77 (ddd,
1, J ) 12.9, 5.5, 5.5), 2.61 (dd, 1, J ) 14.6, 6.3), 2.40 (ddd, 1,
J ) 14.6, 6.5, 0.8), 2.34 (ddd, 1, J ) 15.7, 8.8, 1.1), 2.06 (s, 3),
1.99 (m, 1), 1.79-1.51 (m, 6), 1.30 (t, 3, J ) 7.1), 1.23 (m, 1);
13C NMR 201.6, 171.5, 170.7, 145.9, 132.5, 66.4, 61.2, 58.7,
38.4, 35.2, 30.7, 29.2, 27.6, 26.2, 21.0, 14.1; IR (neat) 1736,
The data for 9a were obtained from the mixture: 1H NMR
6.10 (br dd, 1, J ) 9.1, 2.4), 5.65-5.69 (m, 1), 4.25 (q, 2, J )
7.1), 2.95 (br d, 1, J ) 17.0), 2.61-2.49 (m, 3), 2.37 (ddd, 1,
J ) 12.8, 10.3, 2.5), 2.28-2.14 (m, 1), 1.84-1.73 (m, 1), 1.29
(t, 3, J ) 7.1); 13C NMR (partial) 132.0, 125.4, 61.4, 43.3, 42.4,
29.7, 14.2.
Oxid a tive Cycliza tion of Eth yl 2-Oxo-1-(2-p r op en yl)-
cycloh exa n eca r boxyla te (1b). Reaction of â-keto ester 1b
(105 mg, 0.5 mmol), Mn(OAc)3‚2H2O (303 mg, 1.125 mmol),
and Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed
AcOH for 16 h at 80 °C followed by normal workup gave 99
mg of a pale yellow oil. Purification by flash chromatography
(20:1 hexane/EtOAc) afforded 1 mg (<1%) of recovered 1b,
followed by 4 mg (4%) of 10b, 66 mg (63%) of 8b, and 7 mg
(7%) of 9b.
1702; UV (EtOH) λmax 231.1 nm (ꢀ 6667). Anal. Calcd for C16
H22O5: C, 65.29; H, 7.54. Found: C, 65.69; H, 7.50.
-
Oxid a tive Cycliza tion of Eth yl 1-(2(E)-Bu ten yl)-2-oxo-
cyclop en ta n eca r boxyla te (1e). Reaction of 1e (105 mg, 0.5
mmol), Mn(OAc)3‚2H2O (304 mg, 1.125 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 10 mL of degassed AcOH
at 80 °C for 1.5 h followed by normal workup gave 101 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 75 mg (75%) of a 10:1 mixture of 8e
and 9e, followed by 11 mg (11%) of a 2:1 mixture of 8e and
9e, 1 mg (1%) of enone 6e, and 3 mg (2%) of acetate 7e. The
calculated yields are <1% of enone 6e, 2% of acetate 7e, 73%
of 8e, and 10% of 9e.
1
The H NMR data for 6e: 1H NMR 7.77-7.31 (m, 1), 6.19-
The data for 8b: 1H NMR 5.96 (dt, 1, J ) 9.6, 3.5), 5.58
(dddd, 1, J ) 9.6, 6.0, 2.4, 1.5), 4.23 (q, 2, J ) 7.2), 3.37 (br d,
1, J ) 18.6), 2.95-2.91 (m, 1), 2.56 (ddd, 1, J ) 18.6, 3.5, 1.5),
2.38-2.32 (m, 1), 2.07-1.92 (m, 4), 1.69-1.63 (m, 1), 1.29 (t,
3, J ) 7.2); 13C NMR 210.3, 171.9, 128.9, 126.1, 61.3, 58.4,
47.4, 39.2, 38.9, 32.9, 17.1, 14.1; IR (neat) 1738, 1716. The
data are identical to those previously reported.6,7
6.13 (m, 1), 5.58-5.50 (m, 2), 4.17 (q, 2, J ) 7.1), 3.24-3.14
(m, 1), 2.69-2.54 (m, 3), 1.64-1.61 (m, 3), 1.24 (t, 3, J ) 7.1).
The data for 8e: 1H NMR 5.28 (m, 1), 4.22 (q, 2, J ) 7.1),
3.18 (dd, 1, J ) 17.1, 2.2), 2.66-2.63 (m, 1), 2.56 (ddd, 1, J )
17.1, 3.7, 2.0), 2.45-2.42 (m, 1), 2.13-2.05 (m, 3), 1.74 (d, 3, J
) 1.5), 1.28 (t, 3, J ) 7.1); 13C NMR 211.4, 171.4, 140.2, 118.3,
61.3, 55.1, 50.9, 42.9, 30.8, 27.9, 22.0, 14.2; IR (neat) 1760,
1727. The data are identical to those previously reported.13
Partial 1H NMR data for 9e were obtained from the 2:1
mixture of 8e and 9e: 6.01 (dd, 1, J ) 9.3, 2.6), 5.45 (ddd, 1,
J ) 9.3, 2.1, 1.2), 4.24 (q, 2, J ) 7.1), 1.29 (t, 3, J ) 7.1), 1.10
(d, 3, J ) 7.2).
The data for 9b: 1H NMR 6.06 (br dt, 1, J ) 9.7, 3.2), 5.66
(ddd, 1, J ) 9.7, 2.6, 1.3), 4.25 (q, 2, J ) 7.2), 2.82 (dddd, 1, J
) 18.6, 7.2, 3.2, 2.6), 2.68-2.65 (m, 1), 2.46 (ddd, 1, J ) 18.6,
3.2, 1.3), 2.40-2.28 (m, 1), 2.05-1.86 (m, 4), 1.17-1.65 (m, 1)
1.28 (t, 3, J ) 7.2); IR (neat) 1741, 1714. The data are identical
to those previously reported.6
Oxid a tive Cycliza tion of Eth yl 1-(2(Z)-bu ten yl)-2-oxo-
cyclop en ta n eca r boxyla te (1f). Reaction of 1f (105 mg, 0.5
mmol), Mn(OAc)3‚2H2O (402 mg, 1.5 mmol), and Cu(OAc)2‚H2O
(100 mg, 0.5 mmol) in 10 mL of degassed AcOH at 80 °C for 2
h followed by normal workup gave 112 mg of a pale yellow
oil. Purification by flash chromatography (20:1 hexane/EtOAc)
afforded 13 mg (13%) of 8f ) 8e, followed by 4 mg (4%) of enone
6f, 8 mg (6%) of a 1:1 mixture of acetates 7f, and 65 mg (65%
by weight) of polymeric material.
The data for 10b: 1H NMR 5.05 (br s, 2), 4.22 (q, 2, J )
7.1), 3.29 (br dd, 1, J ) 17.3, 1.2), 2.98-2.96 (m, 1), 2.73 (br
dd, 1, J ) 17.3, 1.1), 2.36 (dddd, 1, J ) 15.1, 12.3, 5.5, 1.5),
2.20-1.97 (m, 3), 1.94-1.76 (m, 1), 1.68-1.61 (m, 1), 1.26 (t,
3, J ) 7.1). The data are identical to those previously
reported.8
Oxid a tive Cycliza tion of Meth yl 2-Oxo-1-(2-p r op en yl)-
cycloh ep ta n eca r boxyla te (1c). Mn(OAc)3‚2H2O (1.333 g,
4.97 mmol), Cu(OAc)2 (225 mg, 1.24 mmol), and 1c (261 mg,
1.24 mmol) were stirred in 12 mL of glacial acetic acid at 80
°C for 85 h. Normal workup and flash chromatography of the
residue on silica gel (10:1 hexane/EtOAc, 2:1:0.2 hexane/
EtOAc/CH2Cl2) gave 50 mg (19%) of unreacted 1c, followed
by 36 mg of a 1:5 mixture of 1c and 10c, 79 mg of a 1:2 mixture
of 10c and 8c, and 4 mg of a 1:3:7 mixture of 10c, 8c, and 9c.
The calculated yields of 8c, 9c, and 10c are 21%, 1%, and 22%,
respectively.
The data for 6f: 1H NMR 7.76 (dt, 1, J ) 5.7, 2.7), 6.17 (dt,
1, J ) 5.7, 2.7), 5.59 (dddq, 1, J ) 8.4, 1.2, 1.2, 6.9), 5.21 (ddt,
1, J ) 8.4, 1.4, 7.2), 4.16 (q, 2, J ) 7.1), 3.02 (br dt, 1, J )
19.3, 2.5), 2.78 (br dd, 1, J ) 14.4, 7.2), 2.60 (ddd, 1, J ) 19.3,
2.7, 2.2), 2.54 (br dd, 1, J ) 14.4, 7.2), 1.63 (d, 3, J ) 6.9), 1.24
(t, 3, J ) 7.1).
1
The H NMR data for acetates 7f were determined from the
1:1 mixture: 5.71-5.60 (m, 1), 5.38-5.30 (m, 1), 5.34 (dd, 0.5
× 1, J ) 11.5, 8.7), 5.08 (dd, 0.5 × 1, J ) 11.5, 8.5), 4.19 (q,
0.5 × 2, J ) 7.1), 4.17 (q, 0.5 × 2, J ) 7.1), 2.72 (br dd, 0.5 ×
1, J ) 13.2, 7.2), 2.70 (br dd, 0.5 × 1, J ) 13.2, 7.2), 2.56-2.35
(m, 2), 2.15 (s, 0.5 × 3), 2.14 (s, 0.5 × 3), 2.14-1.95 (m, 2),
1.88-1.70 (m, 0.5), 1.67-1.62 (m, 0.5), 1.66 (d, 0.5 × 3, J )
6.9), 1.62 (d, 0.5 × 3, J ) 6.9), 1.25 (t, 0.5 × 3, J ) 7.1), 1.27
(t, 0.5 × 3, J ) 7.1).
The data for 8c: 1H NMR 5.84 (br dd, 1, J ) 9.8, 6.0), 5.66
(br dd, 1, J ) 9.8, 2.8), 3.78 (s, 3), 3.06 (m, 2), 2.50 (br dd, 1,
J ) 13.9, 6.6), 2.39 (ddd, 1, J ) 17.0, 6.0, 1.0), 1.90-1.32 (m,
6), 1.22 (m, 1); 13C NMR 212.0, 172.3, 129.1, 125.9, 61.5, 52.3,
50.7, 37.0, 33.3, 31.5, 27.4, 26.1; IR (neat) 1738, 1716. Anal.
Calcd for C12H16O3 (for the 2:1 mixture of 8c and 10c): C,
69.21; H, 7.74. Found: C, 69.37; H, 7.91.
A trace of 9f (<1%) was observed in the crude mixture of
products: 1H NMR 6.05 (d, 1, J ) 9.9), 1.11 (d, 3, J ) 7.1).
Oxid a tive Cycliza tion of Eth yl 1-(2(E)-Bu ten yl)-2-oxo-
cycloh exa n eca r boxyla te (12a ). Reaction of 12a (112 mg,
0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 10 mL of degassed AcOH
at 80 °C for 16 h followed by normal workup gave 107 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 38 mg (35%) of a 1:1:5:2 mixture of
16, 18, 20, and 24a , followed by 42 mg (38%) of a 5.6:3.2:7:2:
1:2 mixture of 16, 18, 20, 21a , 23a , and 24a and 8 mg (7%) of
a 2:1:3:1 mixture of 16, 18, 21a , and 23a . The calculated yields
of the products are 18% of 16, 13% of 18, 32% of 20, 7% of
21a , 2% of 23a , and 7% of 24a .
The data for 9c: 1H NMR 6.04 (br dd, 1, J ) 9.8, 6.1), 5.57
(dd, 1, J ) 9.8, 2.9), 3.68 (s, 3), 2.88 (br ddd, 1, J ) 10.7, 4.9,
4.2), 2.70 (dddd, 1, J ) 17.5, 4.9, 2.9, 2.4), 2.30 (ddd, 1, J )
17.5, 6.1, 1.7), 2.16 (m, 2), 2.00-1.40 (m, 4), 1.13 (m, 2); 13C
NMR 210.2, 172.0, 128.5, 128.4, 62.2, 52.5, 46.2, 34.0, 33.2,
30.3, 25.9, 25.2; IR (neat) 1742, 1710.
The data for 10c: 1H NMR 5.12 (br s, 1), 5.06 (br s, 1), 3.76
(s, 3), 3.42 (br d, 1, J ) 17.2), 3.14 (m, 1), 2.64 (dddd, 1, J )
17.2, 2, 2, 2), 2.16-1.97 (m, 2), 1.85-1.54 (m, 4), 1.39 (m, 2);
13C NMR 213.5, 172.5, 147.7, 109.2, 61.5, 52.6, 51.5, 40.5, 34.5,
31.3, 25.0, 23.1; IR (neat) 1755, 1732.
Oxid a tive Cycliza tion of Eth yl 2-Oxo-1-(2-p r op en yl)-
cycloocta n eca r boxyla te (1d ). Mn(OAc)3‚2H2O (2.03 g, 7.56
mmol), Cu(OAc)2 (171 mg, 0.95 mmol), and 1d (226 mg, 0.95
mmol) were stirred in 20 mL of glacial acetic acid at 80 °C for
226 h. Normal workup and flash chromatography of the
residue on silica gel (15:1 hexane/EtOAc, 2:1:0.2 hexane/
Reaction of 12a (112 mg, 0.5 mmol), Mn(OAc)3‚2H2O (402
mg, 1.25 mmol), and Cu(OAc)2‚H2O (10 mg, 0.05 mmol) in 10
mL of degassed AcOH at 80 °C for 16 h followed by normal
workup gave 109 mg of a pale yellow oil. Purification by flash

7842 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
chromatography (20:1 hexane/EtOAc) afforded 5 mg (4%) of
recovered 12a , followed by 21 mg (19%) of a 5:1 mixture of
24a and 20, 27 mg (25%) of a 0.1:2:10:3 mixture of 16, 18, 20,
and 24a , 22 mg (20%) of a 3:1:3 mixture of 16, 18, and 20, 10
mg (9%) of a 3:1:3:3 mixture of 16, 18, 21a , and 23a , 6 mg
(5%) of a 3:1:6:3 mixture of 16, 18, 21a , and 23a , and 2 mg
(2%) of a 1:2:1 mixture of 16, 21a , and 23a . The calculated
yields of the products are 15% of 16, 10% of 18, 26% of 20, 6%
of 21a , 5% of 23a , and 18% of 24a .
The data for 16: 1H NMR 5.79 (ddd, 1, J ) 17.0, 10.1, 8.3),
5.03 (br d, 1, J ) 17.0), 4.95 (br d, 1, J ) 10.1), 4.23 (q, 2, J )
7.1), 2.66 (ddd, 1, J ) 9.1, 8.3, 4.6), 2.42-2.27 (m, 4), 2.25 (ddd,
1, J ) 11.7, 5.3, 1.2), 2.15-2.06 (m, 1), 1.92-1.79 (m, 2), 1.75-
1.67 (m, 1), 1.28 (t, 3, J ) 7.1); 13C NMR 214.4, 171.1, 141.5,
113.3, 61.2, 58.4, 52.4, 38.8, 38.4, 36.1, 34.5, 18.1, 14.2; IR
(neat) 1749, 1728.
Partial 1H NMR data for 18 were obtained from the
mixture: 6.01 (dddd,1, J ) 16.6, 10.1, 6.8, 0.8), 5.40 (br d, 1,
J ) 16.6), 5.25 (br d, 1, J ) 10.1).
IR (neat) 1757, 1728. Anal. Calcd for C13H18O3: C, 70.24; H,
8.16. Found: C, 69.91; H, 8.27.
The data for 27a : 1H NMR 7.75 (dt, 1, J ) 5.8, 2.7), 6.16
(dt, 1, J ) 5.8, 2.2), 4.94 (dt, 1, J ) 1.5, 7.8), 4.16 (q, 2, J )
7.1), 3.19 (ddd, 1, J ) 19.2, 2.7, 2.2), 2.72 (br dd, 1, J ) 14.5,
7.8), 2.59 (ddd, 1, J ) 19.2, 2.7, 2.2), 2.47 (br dd, 1, J ) 14.5,
7.8), 1.65 (s, 3), 1.63 (s, 3), 1.24 (t, 3, J ) 7.1).
Partial 1H NMR data for 28a were obtained from the
mixture: 4.75 (br s,1), 4.73 (br s,1), 4.21 (q, 2, J ) 7.1), 1.56
(s, 3), 1.25 (t, 3, J ) 7.1).
Oxid a tive Cycliza tion of Eth yl 1-(3-Meth yl-2-bu ten yl)-
2-oxocycloh exa n eca r boxyla te (25b). Reaction of â-keto
ester 25b (112 mg, 0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25
mmol), and Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of
degassed AcOH at 80 °C for 18 h followed by normal workup
gave 106 mg of a pale yellow crude oil. Purification by flash
chromatography (20:1 hexane/EtOAc) afforded 15 mg (13%)
of recovered 25b, followed by 56 mg (51%) of 28b.
A trace of 26b was observed in the 1H NMR spectrum of
recovered 25b: 1H NMR 5.91 (d, 1, J ) 9.8), 1.00 (s, 3), 0.98
(s, 3).
Partial 1H NMR data for 20 were obtained from the
mixture: 5.56 (m, 1), 4.23 (q, 2, J ) 7.2), 3.32 (br d, 1, J )
18.4), 2.74-2.69 (m, 1), 2.44-2.40 (m, 1), 1.29 (t, 3, J ) 7.2).
The data are identical to those previously reported.14
Partial 1H NMR data for 21a were obtained from the
mixture: 5.89 (ddd, 1, J ) 9.7, 2.3, 0.8), 5.58 (ddd, 1, J ) 9.7,
2.7, 1), 4.25 (q, 2, J ) 7.2), 2.99-2.90 (m, 1), 2.45-2.54 (m, 1),
1.30 (t, 3, J ) 7.2); 1.16 (d, 3, J ) 7.5).
The data for 28b: 1H NMR 4.75 (t, 1, J ) 0.9), 4.67 (t, 1, J
) 1.5), 4.21 (q, 2, J ) 7.1), 2.67 (dd, 1, J ) 10.3, 4.4), 2.45
(ddd, 1, J ) 14.1, 4.4, 1.4), 2.38-2.22 (m, 3), 2.13 (br dd, 1, J
) 13.3, 5.0), 2.06-2.00 (m, 2), 1.94-1.72 (m, 2), 1.69 (d, 3, J
) 1.4), 1.28 (t, 3, J ) 7.1); 13C NMR (partial) 171.2, 110.1,
61.1, 58.7, 51.9, 42.5, 38.6, 36.6, 34.1, 19.7, 18.0, 14.1; IR (neat)
1748, 1727.
Partial 1H NMR data for 23a were obtained from the
mixture: 6.00 (dt, 1, J ) 9.8, 3.6), 5.64 (ddd, 1, J ) 9.8, 2.2,
1.5), 4.22 (q, 2, J ) 7.1), 3.65 (br d, 1, J ) 17.7), 1.30 (t, 3, J
) 7.1), 1.02 (d, 3, J ) 6.8).
A trace of 29b was observed in the 1H NMR spectrum of
28b: 5.11 (br s, 1), 5.01 (br s, 1).
Oxid a tive Cycliza tion of Eth yl 1-(2-Meth yl-2-p r op en -
yl)-2-oxocyclop en ta n eca r boxyla te (30a ). Reaction of 30a
(105 mg, 0.5 mmol), Mn(OAc)3‚2H2O (302 mg, 1.125 mmol),
and Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed
AcOH at 80 °C for 1.5 h followed by normal workup gave 101
mg of a pale yellow oil. Purification by flash chromatography
(20:1 hexane/EtOAc) afforded 26 mg (26%) of a 3:4:1 mixture
of 35, 34, and 33, followed by 15 mg (15%) of a 2:1.5:1 mixture
of 35, 34, and 33, 9 mg (9%) of a 1:1:2 mixture of 35, 34, and
33, 3 mg (3%) of 44a , 5 mg (5%) of acetate 36a , 5 mg (5%) of
acetate 37a , and 5 mg (5%) of 42a . The calculated yields are
6% of 33a , 22% of 34a , 22% of 35a , 5% of 36a , 5% of 37a , 5%
of 42a , and 3% of 44a . Similar results were obtained from
the oxidative cyclization of 30a with Mn(OAc)3‚2H2O and no
Cu(OAc)2‚H2O.
Partial 1H NMR data for 24a were obtained from the
mixture: 5.98 (ddd, 1, J ) 9.8, 3.3, 3.3), 5.60 (m, 1), 4.23 (q, 2,
J ) 7.2), 3.41 (br d, 1, J ) 18.8), 2.78 (dd, 1, J ) 5.8, 3.6), 2.51
(ddd, 1, J ) 18.8, 3.4, 1.8), 2.34 (dddd, 1, J ) 13.6, 13.6, 5.3,
1.8), 2.16-2.04 (m, 1), 1.90-2.00 (m, 2), 1.76-1.56 (m, 1), 1.29
(t, 3, J ) 7.2), 0.95 (d, 3, J ) 6.8).
Oxid a tive Cycliza tion of Eth yl 1-(2(Z)-Bu ten yl)-2-oxo-
cycloh exa n eca r boxyla te (12b). Reaction of 12b (112 mg,
0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 10 mL of degassed AcOH
at 80 °C for 16 h followed by normal workup gave 108 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 3 mg (3%) of a 1:4 mixture of 24b and
17, followed by 5 mg (4%) of a 1:3:3:2 mixture of 16, 18, 20,
and 24b, 43 mg (40%) of a 12:1 mixture of 16 and 18, 4 mg
(3%) of a 7:1:2:1 mixture of 16, 18, 21b, and 23b, and 6 mg
(5%) of a 5:1 mixture of 18 and 21b. The calculated yields of
the products are 46% of 16, <1% of 17, 5% of 18, 2% of 20, 2%
of 21b, 1% of 23b and 2% of 24b.
Partial 1H NMR data for 33a were obtained from the
mixture: 5.54 (dt, 1, J ) 8.9, 1.7), 4.22 (q, 2, J ) 7.2), 3.25 (br
d, 1, J ) 17.3), 2.58 (d, 1, J ) 17.3), 1.69 (br s, 3), 1.29 (t, 3, J
) 7.2).
Partial 1H NMR data for 34a were obtained from the
mixture: 5.81 (br s, 1), 4.22 (q, 2, J )7.2), 2.89 (br d, 1, J )
16.8), 2.32 (ddd, 1, J ) 12.9, 10.2, 2.5), 1.29 (t, 3, J )7.2).
Partial 1H NMR data for 35a were obtained from the
mixture: 5.03 (br d, 2, J ) 1.9), 4.22 (q, 2, J ) 7.1), 3.01 (br d,
1, J ) 14.1), 2.73 (br d, 1, J ) 14.1), 1.28 (t, 3, J ) 7.1).
The data for 36a : 1H NMR 4.21 (q, 2, J ) 7.1), 3.01 (br dd,
1, J ) 15.2, 4.2), 2.93 (br dt, 1, J ) 15.2, 4.1), 2.64-2.52 (m,
1), 2.48 (dd, 1, J ) 15.4, 1.5), 2.40-2.28 (m, 2), 2.13 (dd, 1, J
) 14.9, 2.8), 2.07 (s, 3), 2.05-1.93 (m, 2), 1.59 (s, 3), 1.28 (t, 3,
J ) 7.1); 13C NMR (partial) 170.8, 170.3, 80.2, 61.4, 55.8, 48.3,
45.7, 44.3, 27.2, 26.3, 22.9, 21.0, 14.2; IR (neat) 1760, 1732.
The data for 37a : 1H NMR 4.22 (q, 2, J ) 7.1), 3.44 (dt, 1,
J ) 15.1, 1.9), 2.82 (dddd, 1, J ) 15.7, 6.0, 2.3, 1.2), 2.56 (ddd,
1, J ) 11.3, 9.8, 1.5), 2.38-2.31 (m, 2), 2.26 (dd, 1, J ) 14.7,
1.0), 2.21-1.90 (m, 2), 1.93 (s, 3), 1.83-1.69 (m, 1), 1.69 (s, 3),
1.29 (t, 3, J ) 7.1); 13C NMR 211.3, 170.9, 170.1, 77.7, 61.4,
53.8, 49.3, 48.5, 40.5, 28.9, 26.5, 24.1, 22.4, 14.2; IR (neat) 1755,
1732. Anal. Calcd for C14H20O5 (1:1 mixture of acetates 36a
and 37a ): C, 62.67; H, 7.51. Found: C, 62.65; H, 7.29.
The data for 42a : 1H NMR 6.91 (d, 1, J ) 2.4), 4.21 (q, 2, J
) 7.1), 3.05 (dt, 1, J ) 2.4, 6.9), 2.92 (dt, 1, J ) 17.1, 4.4), 2.56
(dddd, 1, J ) 17.7, 12.2, 4.2, 2.4), 2.33-2.17 (m, 3), 1.86 (dddd,
1, J ) 14.2, 12.2, 3.9, 2.9), 1.64 (s, 3), 1.30 (t, 3, J ) 7.1); 13C
NMR 179.4, 167.7, 144.2, 136.2, 83.9, 61.8, 42.7, 42.6, 30.7,
28.4, 24.3, 14.6; IR (neat) 1776, 1710. Anal. Calcd for
C14H16O4: C, 64.27; H, 7.19. Found: C, 64.45; H, 6.90.
Partial 1H NMR data for 17 were obtained from the
mixture: 5.50-5.38 (m, 1), 1.66 (br d, 3, J ) 8.8).
Partial 1H NMR data for 21b were obtained from the
mixture: 6.02-5.98 (m, 1), 5.61 (dd, 1, J ) 9.6, 1.2), 1.07 (d,
3, J ) 7.2).
Partial 1H NMR data for 23b were obtained from the
mixture: 5.98-5.94 (m, 1), 5.68-5.63 (m, 1), 1.00 (d, 3, J )
7.3).
Partial 1H NMR data for 24b were obtained from the
mixture: 5.90 (dt, 1, J ) 9.4, 3.5), 5.68-5.62 (m, 1), 4.23 (q, 2,
J ) 7.1), 3.42 (br d, 1, J ) 17.9), 2.66-2.52 (m, 3), 1.27 (t, 3,
J ) 7.1), 1.05 (d, 3, J ) 7.6).
Oxid a tive Cycliza tion of Eth yl 1-(3-Meth yl-2-bu ten yl)-
2-oxocyclop en ta n eca r boxyla te (25). Reaction of 25 (120
mg, 0.5 mmol), Mn(OAc)3‚2H2O (402 mg, 1.5 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 10 mL of degassed AcOH
at 80 °C for 2.5 h followed by normal workup gave 108 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 6 mg (5%) of 26a , followed by 7 mg
(6%) of a 2:1 mixture of 26a and 28a , 7 mg (6%) of enone 27a ,
and 72 mg of polymeric material.
The data for 26a : 1H NMR 5.93 (d, 1, J ) 9.3), 5.39 (dd, 1,
J ) 9.3, 1.4), 4.25 (q, 2, J ) 7.1), 2.53-2.43 (m, 1), 2.17-1.97
(m, 4), 1.30 (t, 3, J ) 7.1), 1.12 (br s, 6); 13C NMR 209.1, 170.5,
136.0, 127.5, 61.3, 54.4, 47.3, 31.9, 27.8, 26.0 (2C), 18.8, 14.2;

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7843
The data for 44a : 1H NMR 5.89 (ddd, 1, J ) 11.3, 7.4, 4.0),
5.74 (ddd, 1, J ) 11.3, 1.8, 0.7), 4.22 (q, 2, J ) 7.1), 2.79 (ddd,
1, J ) 12.3, 1.7, 1.0), 2.57 (dddd, 1, J ) 13.8, 4.4, 4.4, 1.6),
2.46-2.33 (m, 2), 2.35 (d, 1, J ) 12.3), 2.06 (ddt, 1, J ) 13.8,
11.3, 4.8), 1.63 (s, 3), 1.29 (t, 3, J ) 7.1); 13C NMR (partial)
135.4, 132.8, 82.8, 62.1, 56.9, 47.0, 33.6, 27.8, 24.5, 14.0; IR
(neat) 1769, 1737. Anal. Calcd for C12H16O4: C, 64.27; H, 7.19.
Found: C, 64.28; H, 6.91.
acetates 54, and 5 mg (7%) of R-hydroxy ketones 55. The
calculated yields are 44% of 52, 7% of 53, 10% of 54, and 7%
of 55.
The data for 52:
1H NMR 5.84 (dddd, 1, J ) 9.0, 6.7, 2.3,
1.0), 5.55 (ddd, 1, J ) 9.0, 3.6, 2.7), 2.56-2.51 (m, 2), 2.05-
1.88 (m, 4), 1.43-1.18 (m, 23), 0.88 (t, 3, J ) 6.6); 13C NMR
219.3, 132.8, 126.0, 48.9, 46.5, 33.2, 31.9, 31.6, 30.7, 30.4, 29.7,
29.6 (2C), 29.5, 29.3, 28.7, 24.2, 24.1, 22.7, 14.1; IR (neat) 1741.
Anal. Calcd for C20H34O: C, 82.69; H, 11.80. Found: C, 82.88;
H, 11.32.
Oxid a tive Cycliza tion of Eth yl 1-(2-Meth yl-2-bu ten yl)-
2-oxocycloh exa n eca r boxyla te (30b). Reaction of â-keto
ester 30b (112 mg, 0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25
mmol), and Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of
degassed AcOH at 80 °C for 16 h followed by normal workup
gave 108 mg of a pale yellow crude oil. Purification by flash
chromatography (20:1 hexane/EtOAc) afforded 2 mg (2%) of
recovered 30b, followed by 48 mg (40%) of a 2.25:1 mixture of
34b and 35b, 18 mg (15%) of a 7:3:1 mixture of 33b, 34b, and
35b, 21 mg (17%) of 33b, and 3 mg (2%) of ethyl 6-acetoxy-1-
(2-methyl-2-butenyl)-2-oxocyclohexanecarboxylate. The cal-
culated yields are 27% of 33b, 32% of 34b, and 13% of 35b.
The data for 33b: 1H NMR 5.38 (d, 1, J ) 0.9), 4.24 (q, 2, J
) 7.1), 2.78-2.61 (m, 2), 2.37-2.27 (m, 2), 1.96-1.78 (m, 4),
1.81 (s, 3), 1.66-1.61 (m, 1), 1.29 (t, 3, J ) 7.1); IR (neat) 1748,
1706. The data are identical to those previously reported.19
Partial 1H NMR data for 34b were obtained from the
mixture: 1H NMR 5.30 (br dq, 1, J ) 6.1, 1.5), 4.23 (q, 2, J )
7.1), 3.33 (br d, 1, J ) 18.3), 2.44 (d, 1, J ) 18.3), 1.78 (d, 3, J
) 1.3), 1.30 (t, 3, J ) 7.1). The data are identical to those
previously reported.19
Partial 1H NMR data for 53 were obtained from the
mixture: 5.66 (dd, 1, J ) 8.8, 1.7).
The data for acetates 54: 1H NMR 6.0-5.9 (m, 1), 5.8-5.5
(m, 1), 5.1-5.0 (m, 2), 1.6-2.5 (m, 6), 2.14 (0.4 × 3, s), 2.135
(0.6 × 3, s), 1.0-1.4 (m, 22), 0.88 (t, 3, J ) 6.6).
The data for alcohols 55: 1H NMR 5.8-5.5 (m, 1), 5.1-5.0
(m, 2), 4.9-4.8 (m, 1), 1.6-2.5 (m, 6), 1.0-1.4 (m, 22), 0.88 (t,
3, J ) 6.6).
Oxid a tive Cycliza tion of 2-(2-P r op en yl)cycloh exa n on e
(56a ). Reaction of 56a (138 mg, 1.0 mmol), Mn(OAc)3‚2H2O
(1.073 g, 4.0 mmol), KOAc (496 mg, 5.0 mmol), and Cu-
(OAc)2‚H2O (200 mg, 1.0 mmol) in 10 mL of degassed AcOH
at 80 °C for 24 h followed by normal workup gave 146 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 9 mg (6%) of recovered 56a , followed
by 43 mg (32%) of 57a , 30 mg (22%) of a 20:1:1 mixture of
57a , 58a , and 61a , 10 mg (7%) of a 1:1 mixture of acetates
59a and 60a , and 5 mg (4%) of acetate 62a . The calculated
yields are 51% of 57a , 1% of 58a , 3% of 59a , 3% of 60a , 1% of
61a , and 3% of 62a .
Partial 1H NMR data for 35b were obtained from the
mixture: 4.90 (t, 2, J ) 1.9), 4.22 (q, 2, J ) 7.1), 1.29 (t, 3, J
) 7.1).
The data for 57a : 1H NMR 5.94 (dt, 1, J ) 9.5, 3.4), 5.58
(dddd, 1, J ) 9.5, 5.9, 2.4, 1.5), 2.85-2.82 (m, 1), 2.75 (br dd,
1, J ) 18.6, 7.3), 2.61-2.56 (m, 1), 2.46 (br dd, 1, J ) 18.6,
3.4), 2.04-1.84 (m, 5), 1.57-1.51 (m, 1); 13C NMR 216.6, 129.9,
127.2, 47.7, 45.5, 36.9 (2C), 33.2, 16.9; IR (neat) 1729. The
data are identical to those previously reported.14
Oxid a tive Cycliza tion of Eth yl 3-Meth yl-2-oxo-1-(2-
p r op en yl)cyclop en ta n eca r boxyla te (45). Reaction of 45
(105 mg, 0.5 mmol), Mn(OAc)3‚2H2O (402 mg, 1.5 mmol), and
Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed AcOH
at 80 °C for 20 h followed by normal workup gave 104 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 32 mg (31%) of 46, followed by 20 mg
(19%) of a 1:1 mixture of 46 and 47, 3 mg (3%) of acetates 48,
and 25 mg (24%) of a 2:3 mixture of lactones 50 and 49. The
calculated yields are 41% of 46, 9% of 47, 3% of 48, 9% of 50,
and 15% of 49.
Partial 1H NMR data for enone 58a were obtained from the
mixture: 6.94 (dt, 1, J ) 3.5, 10.1), 5.89-5.66 (m, 2), 5.11-
4.96 (m, 2).
Partial 1H NMR data for enone 61a 20 were obtained from
the mixture: 6.73 (t, 1, J ) 4.2), 5.89-5.66 (m, 1), 5.11-4.96
(m, 2).
The data for 62a : 1H NMR 5.84 (dt, 1, J ) 9.5, 3.6), 5.57
(dddd, 1, J ) 9.5, 5.9, 2.4, 1.7), 3.16 (br d, 1, J ) 18.1), 3.13
(m, 1), 2.92 (ddd, 1, J ) 18.1, 3.6, 1.7), 2.18-2.00 (m, 3), 2.11
(s, 3), 1.84-1.89 (m, 2), 1.73-1.65 (m, 1); 13C NMR 206.5,
169.2, 127.9, 125.9, 83.9, 49.2, 42.1, 42.0, 33.2, 21.7, 17.9; IR
(neat) 1753, 1731.
The data for 46: 1H NMR 5.58 (ddd, 1, J ) 9.1, 3.8, 2.2),
5.52 (br d, 1, J ) 9.1), 4.23 (q, 2, J ) 7.2), 3.23 (br d, 1, J )
17.4), 2.68 (dd, 1, J ) 17.4, 3.8), 2.62-2.56 (m, 1), 2.22 (ddd,
1, J ) 13.6, 10.1, 1.9), 2.06 (ddd, 1, J ) 13.2, 10.1, 7.8), 1.87-
1.77 (m, 1), 1.29 (t, 3, J ) 7.2), 1.11 (s, 3); 13C NMR 211.8,
171.5, 137.4, 124.6, 61.2, 56.4, 48.0, 44.4, 36.5, 29.9, 16.6, 14.2;
IR (neat) 1760, 1727.
Oxidative Cyclization of 2-(2-P r open yl)cycloh eptan on e
(56b). Mn(OAc)3‚2H2O (1120 mg, 4.18 mmol), Cu(OAc)2 (126
mg, 0.696 mmol), and 56b (106 mg, 0.696 mmol) were stirred
in 8 mL of glacial acetic acid at 80 °C for 60 h. Normal workup
and flash chromatography of the residue on silica gel (24:1
hexane/EtOAc) gave 10 mg (9%) of unreacted 56b, followed
by 33 mg (32%) of 57b and 16 mg (11%) of 60b.
The data for 47 were obtained from the mixture: 1H NMR
6.09 (ddd, 1, J ) 9.3, 2.1, 1.1), 5.63 (ddd, 1, J ) 9.3, 3.7, 2.6),
4.26 (q, 2, J ) 7.1), 3.22 (br d, 1, J ) 17.5), 2.67 (dd, 1, J )
17.5, 3.7), 2.66-2.59 (m, 1), 2.28 (ddd, 1, J ) 12.7, 10.1, 3.1),
2.04 (ddd, 1, J ) 13.1, 10.2, 7.3), 1.94-1.85 (m, 1), 1.30 (t, 3,
J ) 7.1), 1.12 (s, 3); 13C NMR 212.2, 170.4, 132.1, 124.7, 61.3,
58.9, 49.9, 46.2, 33.1, 32.3, 19.5, 14.2.
The data for 57b: 1H NMR 5.83 (m, 1), 5.63 (ddd, 1, J )
9.8, 3.1, 3.0), 2.95 (m, 1), 2.74 (dddd, 1, J ) 10.2, 5.5, 5.1, 1.3),
2.56 (dddd, 1, J ) 17.5, 5.1, 5.1, 3.0), 2.25 (ddd, 1, J ) 17.5,
5.5, 1.8), 2.12 (m, 1), 1.81-1.65 (m, 3), 1.52 (m, 2), 1.24 (dddd,
1, J ) 13.5, 13.5, 5.9, 1.5), 1.02 (m, 1); 13C NMR 209.0, 129.5,
126.3, 49.9, 46.6, 34.5, 31.0, 30.4, 27.1, 25.9; IR (neat) 1716.
The data for 60b: 1H NMR 5.73 (dddd, 1, J ) 18.0, 9.4, 7.6,
6.9), 5.12 (br d, 1, J ) 18.0), 5.09 (br d, 1, J ) 9.4), 2.91 (dd,
1, J ) 14.8, 6.9), 2.73 (ddd, 1, J ) 15.1, 6.4, 6.4), 2.59 (dd, 1,
J ) 14.8, 7.6), 2.44 (ddd, 1, J ) 15.1, 6.7, 6.7), 2.24 (ddd, 1, J
) 14.6, 6.6, 3.0), 2.09 (s, 3), 1.77-1.35 (m, 7); 13C NMR 207.2,
170.1, 132.3, 119.0, 87.8, 40.4, 38.9, 34.6, 28.2, 23.8, 23.1, 20.1;
IR (neat) 1738, 1708.
Oxid a tive Cycliza tion of 2-(2-P r op en yl)cycloocta n on e
(56c). Mn(OAc)3‚2H2O (582 mg, 2.17 mmol), Cu(OAc)2 (66 mg,
0.36 mmol), and 56c (60 mg, 0.36 mmol) were stirred in 8 mL
of glacial acetic acid at 80 °C for 42 h. Normal workup and
flash chromatography of the residue on silica gel (15:1 hexane/
EtOAc) gave 18 mg (30%) of unreacted 56c, followed by 20
mg (25%) of 63c: 1H NMR 6.46 (d, 1, J ) 5.4), 5.37 (dt, 1, J )
5.4, 4.7), 2.76 (m, 2), 2.23 (ddd, 1, J ) 14.6, 7.0, 4.2), 2.14-
The data for acetates 48: 1H NMR 5.65-5.53 (m, 1), 5.17-
5.03 (m, 2), 4.16 (q, 0.5 × 2, J ) 7.1), 4.12 (q, 0.5 × 2, J ) 7.1),
3.09-2.91 (m, 1), 2.79-2.69 (m, 1), 2.59-2.28 (m, 3), 2.05 (s,
0.5 × 3), 2.04 (s, 0.5 × 3), 1.91-1.81 (m, 1), 1.56 (s, 0.5 × 3),
1.41 (s, 0.5 × 3), 1.25 (t, 3, J ) 7.1).
Partial 1H NMR data for 49 were obtained from the
mixture: 5.82-5.73 (m, 2), 4.25 (q, 2, J ) 7.1), 1.49 (s, 3), 1.31
(t, 3, J ) 7.2); IR (neat) of the mixture 1773, 1738.
Partial 1H NMR data for 50 were obtained from the
mixture: 5.73 (br d, 1, J ) 6.6), 4.95 (dd, 1, J ) 8.2, 6.6), 4.24
(q, 2, J ) 7.1), 1.79 (dd, 3, J ) 1.4, 1.1), 1.30 (t, 3, J ) 7.1).
Oxid a t ive Cycliza t ion of 2-Dod ecyl-2-(2-p r op en yl)-
cyclop en ta n on e (51). Reaction of 51 (73 mg, 0.25 mmol),
Mn(OAc)3‚2H2O (201 mg, 0.75 mmol), and Cu(OAc)2‚H2O (50
mg, 0.25 mmol) in 2.5 mL of degassed AcOH at 80 °C for 12 h
followed by normal workup gave 76 mg of a pale yellow oil.
Purification by flash chromatography (20:1 hexane/EtOAc)
afforded 23 mg (32%) of a 7:1 mixture of 52 and 53, followed
by 14 mg (19%) of a 5:1 mixture of 52 and 53, 7 mg (10%) of

7844 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
1.91 (m, 2), 1.99 (s, 3), 1.79 (m, 2), 1.67-1.38 (m, 6); decoupled
at 6.46, 5.34 (t, 1, J ) 4.7), 2.23 (dd, 1, J ) 14.5, 6.8); decoupled
at 5.37, 6.51 (br s, 1); 13C NMR 194.7, 170.6, 147.7, 132.1, 66.8,
44.5, 36.6, 31.8, 31.4, 31.2, 27.0, 24.9, 21.0; IR (neat) 1734,
The data for 75a : 1H NMR 6.80 (ddd, 1, J ) 10.3, 3.1, 1.1),
6.11 (dd, 1, J ) 10.3, 1.6), 5.57 (dddd, 1, J ) 7.8, 5.1, 3.1, 1.6),
4.19 (q, 2, J )7.1), 2.69 (dd, 1, J ) 13.6, 7.8), 2.23 (ddd, 1, J )
13.6, 5.1, 1.1), 2.10 (s, 3), 1.45 (s, 3), 1.26 (t, 3, J ) 7.1); 13C
NMR 196.2, 172.0, 170.1, 145.2, 129.8, 65.7, 61.5, 52.6, 37.9,
20.9, 19.9, 14.0; IR (neat) 1742, 1687.
1699; UV (EtOH) λmax 229.8 nm (ꢀ 5029). Anal. Calcd for C13
-
H18O3: C, 70.24; H, 8.16. Found: C, 70.26; H, 8.41.
Oxid a tive Cycliza tion of 1-Meth yl-1-(2-p r op en yl)-2-
tetr a lon e (64). Reaction of 64 (100 mg, 0.5 mmol), Mn(OAc)3‚
2H2O (335 mg, 1.25 mmol), and Cu(OAc)2‚H2O (100 mg, 0.5
mmol) in 5 mL of degassed AcOH at 80 °C for 16 h followed
by normal workup gave 92 mg of a pale yellow oil. Purification
by flash chromatography (20:1 hexane/EtOAc) afforded 59 mg
(59%) of 65, followed by 16 mg (16%) of a 2:1 mixture of 65
and 66.
The data for 65: 1H NMR 7.33-7.10 (m, 4), 5.70 (dddd, 1,
J ) 9.3, 4.7, 2.0, 1.0), 5.59 (br dt, 1, J ) 9.3, 3.3), 3.39 (dd, 1,
J ) 16.7, 5.2), 3.19-3.17 (m, 1), 3.15 (dd, 1, J ) 16.7, 1.9),
2.62 (d, 2, J ) 3.3), 1.49 (s, 3); 13C NMR 213.6, 144.6, 132.9,
129.8, 128.4, 127.0, 126.7, 126.5, 125.8, 49.0, 48.3, 46.8, 39.0,
19.9; IR (neat) 1728.
The data for 66 were obtained from the mixture: 1H NMR
7.33-7.12 (m, 4), 5.72-5.65 (m, 1), 5.55 (ddd, 1, J ) 9.1, 2.5,
0.8), 3.56 (dd, 1, J ) 17.4, 8.4), 3.21 (dd, 1, J ) 17.4, 2.4), 3.00
(dddd, 1, J ) 8.4, 5.5, 2.4, 1.4), 2.84 (ddddd, 1, J ) 17.8, 5.5,
2.5, 2.4, 0.7), 2.55 (ddd, 1, J ) 17.8, 4.4, 1.4), 1.51 (s, 3); 13C
NMR (partial) 137.2, 134.8, 128.4, 128.3, 126.7, 126.5, 124.8,
124.2, 42.2, 40.0, 39.2, 17.9.
Oxid a tive Cycliza tion of 1,1-Bis(2-p r op en yl)-4a -m eth -
yl-3,4,4a ,5,6,7-h exa h yd r on a p h th a len -2(1H)-on e (67). Mn-
(OAc)3‚2H2O (363 mg, 1.35 mmol), Cu(OAc)2 (109 mg, 0.60
mmol), and 67 (147 mg, 0.60 mmol) were stirred in 5 mL of
glacial acetic acid at 80 °C for 10 h. Normal workup and flash
chromatography of the residue on silica gel (30:1 hexane/
EtOAc) gave 3 mg of 70, followed by 12 mg of a 1:20 mixture
of 70 and 68, followed by 96.5 mg of 68 and 20 mg of a 1:1.2
mixture of 68 and 69. The calculated yields of 68, 69, and 70
are 80%, 7.5%, and 2.5%, respectively.
Oxid a t ive Cycliza t ion of E t h yl 2-(2(E)-Bu t en yl)-2-
m eth yla cetoa ceta te (71b). Reaction of 71b (99 mg, 0.5
mmol), Mn(OAc)3‚2H2O (804 mg, 3 mmol), and Cu(OAc)2‚H2O
(100 mg, 0.5 mmol) in 10 mL of degassed AcOH at 80 °C for
48 h followed by normal workup gave 103 mg of a pale yellow
oil. Purification by flash chromatography (4:1 hexane/EtOAc)
gave 2 mg of recovered â-keto ester 71b (2%), followed by 12
mg (10%) of acetate 75b , 27 mg (20%) of a 1:2 mixture of
acetates 75b and 74b, and 20 mg (17%) of acetate 74b. The
calculated yields are 30% of 74b and 17% of 75b.
The data for 74b: 1H NMR 5.99 (br s, 1), 5.52 (br dd, 1, J
) 6.9, 4.8), 4.16 (q, 2, J ) 6.9), 2.69 (dd, 1, J ) 13.8, 6.9), 2.15
(dd, 1, J ) 13.8, 4.8), 2.11 (s, 3), 1.95 (s, 3), 1.42 (s, 3), 1.25 (t,
3, J ) 6.9); 13C NMR 196.0, 172.4, 170.2, 156.1, 127.8, 68.2,
61.4, 51.9, 38.0, 20.8, 20.4, 20.3, 14.0; IR (neat) 1742, 1682. A
sample of the mixture of acetates was submitted for CH
analysis. Anal. Calcd for C13H18O5: C, 61.40; H, 7.14.
Found: C, 61.64; H, 7.30.
The data for 75b: 1H NMR 5.98 (br s, 1), 5.81 (br dd, 1, J
) 9.6, 5.4), 4.18 (q, 2, J ) 7.1), 2.73 (dd, 1, J ) 13.2, 5.4), 2.13
(s, 3), 1.92 (dd, 1, J ) 13.2, 9.6), 1.91 (s, 3), 1.41 (s, 3), 1.25 (t,
3, J ) 7.1); 13C NMR 194.7, 171.5, 170.0, 158.5, 127.6, 68.8,
61.7, 53.1, 39.4, 21.3, 20.8, 19.7, 14.0; IR (neat) 1743, 1684.
Oxid a tive Cycliza tion of 3-Acetyl-4,5-d ih yd r o-3-(2-p r o-
p en yl)-2(3H)-fu r a n on e (76a ). Reaction of 76a (93 mg, 0.55
mmol), Mn(OAc)3‚2H2O (951 mg, 3.55 mmol), and Cu(OAc)2
(107 mg, 0.59 mmol) in 10 mL of glacial acetic acid at 80 °C
for 40 h followed by normal workup gave 96 mg of crude
product, which was purified by flash chromatography on silica
gel (1:2:0.4 hexane/EtOAc/CH2Cl2) to yield 30 mg (24%) of 77a ,
followed by 48 mg (39%) of 78a .
The data for 77a : 1H NMR 6.99 (ddd, 1, J ) 10.3, 3.0, 1.3),
6.14 (dd, 1, J ) 10.3, 1.8), 6.06 (dddd, 1, J ) 8.2, 5.4, 3.0, 1.8),
4.46 (dt, 1, J ) 8.8, 7.2), 4.39 (dt, 1, J ) 8.8, 3.7), 2.90 (ddd, 1,
J ) 13.1, 7.2, 3.7), 2.72 (ddd, 1, J ) 13.6, 5.4, 1.3), 2.15 (dd, 1,
J ) 13.6, 8.2), 2.14 (m, 1), 2.11 (s, 3); 13C NMR 193.3, 174.1,
169.8, 148.6, 128.8, 66.7, 65.7, 53.8, 36.4, 33.4, 20.9; IR (neat)
1764, 1738, 1684.
The data for 78a : 1H NMR 6.91 (ddd, 1, J ) 10.4, 2.1, 2.0),
6.06 (dd, 1, J ) 10.4, 2.2), 5.65 (dddd, 1, J ) 10.9, 5.3, 2.2,
2.1), 4.43 (m, 1), 4.33 (m, 1), 2.70 (dd, 1, J ) 12.9, 10.9), 2.48-
2.42 (m, 2), 2.29 (ddd, 1, J ) 12.9, 5.3, 2.0), 2.15 (s, 3); 13C
NMR 193.5, 174.8, 170.2, 149.6, 127.9, 66.2, 65.2, 54.1, 36.4,
32.9, 20.9; IR (neat) 1777, 1744, 1679. Anal. Calcd for
C11H12O5: C, 58.78; H, 5.39. Found: C, 58.99; H, 5.58.
Oxid a tive Cycliza tion of 3-Acetyl-3-(2-bu ten yl)-4,5-
d ih yd r o-2(3H)-fu r a n on e (76b). Mn(OAc)3‚2H2O (585 mg,
2.18 mmol), Cu(OAc)2 (66 mg, 0.36 mmol), and 76b (66 mg,
0.36 mmol) were stirred in 10 mL of glacial acetic acid at 80
°C for 24 h. Normal workup and flash chromatography of the
residue on silica gel (1:2:0.4 hexane/EtOAc/CH2Cl2) provided
12 mg of unreacted 76b, followed by 2 mg (2%) of 79b, 15 mg
(17%) of 77b, and 23 mg (27%) of 78b.
The data for 77b: 1H NMR 6.01 (m, 2), 4.44 (dt, 1, J ) 8.8,
7.1), 4.37 (dt, 1, J ) 8.8, 3.4), 2.84 (ddd, 1, J ) 13.0, 7.1, 3.4),
2.70 (dd, 1, J ) 13.8, 5.4), 2.20-2.05 (m, 2), 2.13 (s, 3), 2.01
(d, 3, J ) 0.8); 13C NMR 193.0, 174.6, 169.9, 159.8, 126.7, 68.0,
66.7, 53.4, 36.2, 33.5, 20.9, 20.7; IR (neat) 1765, 1743, 1666.
The data for 78b: 1H NMR 5.93 (dq, 1, J ) 1.5, 1.5), 5.67
(ddd, 1, J ) 10.8, 5.2, 1.9), 4.42 (m, 1), 4.33 (m, 1), 2.66 (dd, 1,
J ) 12.9, 10.8), 2.42 (m, 1), 2.23 (dd, 1, J ) 12.9, 5.1), 2.16 (s,
3), 1.99 (d, 3, J ) 0.9); 13C NMR 193.2, 175.1, 170.3, 161.0,
125.7, 68.2, 65.3, 54.2, 36.4, 33.1, 20.8, 19.8; IR (neat) 1776,
1744, 1666. Anal. Calcd for C12H14O5: C, 60.50; H, 5.92.
Found: C, 60.36; H, 5.61.
The data for 68: 1H NMR 5.90 (dddd, 1, J ) 17.9, 10.3, 7.5,
7.5), 5.72 (dd, 1, J ) 4.1, 4.1), 5.64 (ddd, 1, J ) 8.9, 5.4, 2.9),
5.47 (br dd, 1, J ) 8.9, 4.9), 5.02 (br d, 1, J ) 10.3), 5.01 (br d,
1, J ) 17.9), 2.92 (dd, 1, J ) 9.5, 5.4), 2.70 (dd, 1, J ) 17.3,
4.9), 2.50 (dd, 1, J ) 13.9, 7.5), 2.35 (br d, 1, J ) 17.3), 2.26
(dd, 1, J ) 13.9, 7.5), 2.03 (m, 3), 1.76 (d, 1, J ) 12.8), 1.62-
1.40 (m, 4), 0.94 (s, 3); 13C NMR 218.5, 146.5, 135.0, 132.8,
124.9, 122.8, 117.4, 52.9, 46.1, 46.0, 45.5, 41.0, 40.4, 33.0, 29.2,
25.4, 19.2; IR (neat) 1727, 1650.
Partial data for 69: 1H NMR (CDCl3) 5.45 (dd, 1, J ) 4.7,
2.9), 5.40 (m, 2), 0.90 (s, 3); 1H NMR (C6D6) 6.12 (dddd, 1, J )
17.1, 10.1, 6.9, 6.9), 5.29 (dd, 1, J ) 9.2, 2.8), 5.23 (dd, 1, J )
4.6, 2.7), 5.14 (m, 1), 0.88 (s, 3).
The data for 70: 1H NMR 5.87 (dddd, 1, J ) 17.0, 10.2, 7.3,
7.3), 5.56 (dd, 1, J ) 4.7, 3.2), 5.04 (m, 2), 5.02 (br d, 1, J )
10.2), 4.96 (br d, 1, J ) 17.0), 2.85 (d, 1, J ) 8.0), 2.51-2.33
(m, 3), 2.32 (dd, 1, J ) 12.9, 8.0), 2.02 (m, 2), 1.78 (d, 1, J )
12.9), 1.65-1.48 (m, 4), 1.30 (m, 1), 0.97 (s, 3); 13C NMR 218.4,
150.8, 147.3, 135.3, 122.4, 117.1, 109.3, 55.2, 52.0, 49.6, 45.1,
40.4, 36.7, 32.3, 29.2, 24.9, 18.4; IR (neat) 1758, 1642.
Oxid a tive Cycliza tion of Eth yl 2-Meth yl-2-(2-p r op en -
yl)a cetoa ceta te (71a ). Reaction of 71a (92 mg, 0.5 mmol),
Mn(OAc)3‚2H2O (804 mg, 3 mmol), and Cu(OAc)2‚H2O (100 mg,
0.5 mmol) in 10 mL of degassed AcOH at 80 °C for 48 h
followed by normal workup gave 101 mg of a pale yellow oil.
Purification by flash chromatography (4:1 hexane/EtOAc)
afforded 40 mg (30%) of acetate 75a , followed by 9 mg (7%) of
a 20:1 mixture of acetates 74a and 75a and 31 mg (23%) of
74a . The calculated yields are 30% of 74a and 30% of 75a .
The data for 74a : 1H NMR 6.77 (br ddd, 1, J ) 10.3, 2.1,
0.8), 6.10 (dd, 1, J ) 10.3, 2.2), 5.73 (dddd, 1, J ) 10.0, 5.4,
2.2, 2.1), 4.18 (q, 2, J ) 7.1), 2.77 (ddd, 1, J ) 13.0, 5.4, 2.0),
2.11 (s, 3), 2.04 (dd, 1, J ) 13.0, 10.0), 1.43 (s, 3), 1.24 (t, 3, J
) 7.1); 13C NMR 195.0, 171.4, 170.0, 147.5, 129.8, 66.9, 61.9,
53.2, 39.5, 21.4, 21.0, 14.0; IR (neat) 1745, 1695. A sample of
a 1:1 mixture of acetates 74a and 75a was submitted for CH
analysis. Anal. Calcd for C12H16O5: C, 59.99; H, 6.71.
Found: C, 60.05; H 6.48.
P a r tia l d a ta for 79b: 1H NMR 5.82 (dd, 1, J ) 9.7, 2.0),
5.54 (d, 1, J ) 10.1), 5.50 (dd, 1, J ) 9.7, 2.8), 4.39 (dd, 1, J )
8.7, 5.4), 4.30 (m, 1), 3.03 (ddd, 1, J ) 13.0, 5.5, 5.5), 2.18 (s,
3), 2.20-2.00 (m, 2), 1.30 (d, 3, J ) 7.0).

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7845
Oxid a t ive Cycliza t ion of 1-P h en yl-6(Z)-n on en -1-on e
(80). Mn(OAc)3‚2H2O (439 mg, 1.64 mmol), Cu(OAc)2 (49 mg,
0.27 mmol), and 80 (59 mg, 0.27 mmol) were stirred in 6 mL
of glacial acetic acid at 80 °C for 60 h. Normal workup and
flash chromatography of the residue on silica gel (20:1 hexane/
EtOAc) gave 6 mg (10%) of recovered phenyl ketone 80,
followed by 9 mg (12%) of 83b, 10 mg (13%) of 83a , and 6 mg
(8%) of 88.
Oxid a tive Cycliza tion of 2-(2(E)-Bu ten yl)-2-m eth yl-1,3-
cyclop en t a n ed ion e (89b). Mn(OAc)3‚2H2O (436 mg, 1.62
mmol), Cu(OAc)2 (118 mg, 0.65 mmol), and 89b (108 mg, 0.65
mmol) were stirred in 8 mL of glacial acetic acid at 85 °C for
2 h. Normal workup and flash chromatography of the residue
on silica gel (8:1 hexane/EtOAc and 2:1 hexane/EtOAc) gave
6 mg (6%) of 92b as a 3:2 mixture of E/ Z isomers, followed by
27 mg (25%) of 90b, 22 mg (20%) of a 1:2 mixture of 90b and
91b, and 10 mg (9%) of 91b. The calculated yields of 90b and
91b are 32% and 23%, respectively.
The data for 90b: 1H NMR 5.27 (br s, 1), 2.93 (d, 1, J )
18.2), 2.93 (d, 1, J ) 6.6), 2.74 (dd, 1, J ) 18.2, 6.6), 2.63 (ddq,
1, J ) 16.9, 3.8, 1.9), 2.54 (ddq, 1, J ) 16.9, 2.4, 2.4), 1.82 (br
s, 3), 1.13 (s, 3); 13C NMR 213.9, 211.7, 142.2, 119.0, 56.6, 50.2,
49.2, 45.9, 22.1, 12.4; IR (neat) 1777, 1732.
The data for 91b: 1H NMR 5.61 (br d, 1, J ) 9.7), 5.24 (dd,
1, J ) 9.7, 2.8), 3.20 (m, 1), 3.10 (d, 1, J ) 19.2), 2.82 (dd, 1,
J ) 8.1, 4.0), 2.54 (dd, 1, J ) 19.2, 8.1), 1.22 (s, 3), 1.20 (d, 3,
J ) 8.2); 13C NMR 210.9, 203.8, 133.6, 130.1, 59.9, 50.2, 43.0,
38.5, 16.4, 9.8; IR (neat) 1770, 1728.
The data for 92b: 1H NMR (E-isomer) 7.23 (s, 2), 5.48 (m,
1), 5.07 (m, 1), 2.43 (br d, 2, J ) 7.8), 1.55 (br s, 3), 1.17 (s, 3);
partial data for Z-isomer 7.22 (s, 2), 2.32 (br d, 2, J ) 7.5),
1.56 (br s, 3), 1.13 (s, 3).
Oxid a tive Cycliza tion of 2-Meth yl-2-(2-p r op en yl)-1,3-
cycloh exa n ed ion e (93). Mn(OAc)3‚2H2O (677 mg, 2.53
mmol), Cu(OAc)2 (115 mg, 0.63 mmol), and 93 (105 mg, 0.63
mmol) were stirred in 12 mL of glacial acetic acid at 80 °C for
5 h. Normal workup and flash chromatography of the residue
on silica gel (10:1 hexane/EtOAc) gave 18 mg (17%) of 96: 1H
NMR 7.06 (dd, 1, J ) 9.8, 6.5), 6.46 (d, 1, J ) 9.8), 3.36 (ddd,
1, J ) 6.5, 3.6, 3.3), 2.05-1.92 (m, 3), 1.79-1.57 (m, 3), 1.23
(s, 3); 13C NMR 209.7, 203.4, 146.6, 132.7, 63.2, 49.1, 41.9, 29.9,
17.4, 15.5; IR (neat) 1731, 1674.
The data for 83b: 1H NMR 7.97 (dd, 1, J ) 7.7, 1.5), 7.52
(ddd, 1, J ) 7.6, 7.6, 1.5), 7.36-7.28 (m, 2), 2.85 (ddd, 1, J )
8, 8, 4), 2.66 (ddd, 1, J ) 8, 8, 8), 2.22-1.92 (m, 2), 2.13 (s, 3),
1.78-1.58 (m, 6), 0.86 (t, 3, J ) 7.3); 13C NMR 194.8, 169.7,
144.7, 133.5, 128.6, 128.5, 128.0, 126.8, 89.8, 47.3, 44.3, 37.7,
33.5, 28.5, 22.7, 21.4, 10.8; IR (neat) 1743, 1694.
The data for 83a : 1H NMR 7.83 (dd, 1, J ) 7.8, 1.4), 7.49
(ddd, 1, J ) 7.6, 7.6, 1.4), 7.35-7.26 (m, 2), 3.32 (ddd, 1, J )
10, 4, 4), 2.53 (ddd, 1, J ) 10, 10, 5), 2.22-2.06 (m, 2), 2.12 (s,
3), 1.92-1.75 (m, 6), 0.67 (t, 3, J ) 7.5); 13C NMR 196.1, 169.7,
144.9, 134.2, 132.7, 128.2, 126.5, 126.4, 90.2, 50.5, 39.9, 29.3,
28.1, 23.9, 21.2, 20.8, 8.3; IR (neat) 1744, 1705.
The data for 88: 1H NMR 8.16 (dd, 1, J ) 8.3, 1.6), 7.57-
7.40 (m, 3), 2.87-2.80 (m, 2), 2.71-2.48 (m, 2), 2.12-1.95 (m,
4), 2.01 (s, 3), 0.40 (t, 3, J ) 7.5); 13C NMR 204.1, 168.8, 143.6,
139.4, 134.6, 132.4, 127.9, 126.3, 123.7, 123.5, 75.0, 54.8, 33.7,
32.8, 30.1, 21.9, 6.9; IR (neat) 1742, 1658.
Oxid a tion of (3a r,9â,9a â)-9-Eth yl-1,2,3,3a ,9,9a -h exa h y-
d r o-4H-ben z[f]in d en -4-on e (81a )22 a n d (3a r,9r,9a r)-9-
E t h yl-1,2,3,3a ,9,9a -h e xa h yd r o-4H -b e n z[f]in d e n -4-on e
(81b).22 Reaction of an inseparable 20:1 mixture of 81a and
81b (1 mg, 0.285 mmol) and Mn(OAc)3‚2H2O (172 mg, 0.64
mmol) in 5 mL of glacial acetic acid at 80 °C for 48 h followed
by normal workup and flash chromatography (10:1 hexane/
EtOAc) provided 20 mg (31%) of a 3:2 mixture of 86 and 87,
followed by 7 mg (9%) of 83b, 9 mg (12%) of 83a , and 10 mg
(13%) of 88.
Oxid a tive Cycliza tion of Eth yl 1-(3-Bu ten yl)-2-oxocy-
clop en ta n eca r boxyla te (97a ). Reaction of 97a (105 mg, 0.5
mmol), Mn(OAc)3‚2H2O (335 mg, 1.25 mmol), and Cu(OAc)2‚H2O
(100 mg, 0.5 mmol) in 10 mL of degassed AcOH at 80 °C for
2.5 h followed by normal workup gave 101 mg of a pale yellow
oil. Purification by flash chromatography (20:1 hexane/EtOAc)
afforded 4 mg (4%) of recovered 97a , followed by 22 mg (22%)
of a 1:4.5 mixture of 99a and 98a , 25 mg (25%) of a 1:3.5:8
mixture of 99a , 98a , and 100a , 10 mg (10%) of 100a , and
(elution with 10:1 hexane/EtOAc) 12 mg (12%) of 101a . The
calculated yields are 25% of 98a , 10% of 99a , 25% of 100a ,
and 12% of 101.
The data for 98a were obtained from the mixture: 1H NMR
5.67-5.65 (m, 2), 4.20 (q, 2, J ) 7.1), 2.74-2.71 (m, 1), 2.57-
2.15 (m, 6), 1.97 (ddd, 1, J ) 13.2, 9.7, 6.6), 1.69 (dddd, 1, J )
12.7, 9.9, 4.8, 2.2), 1.27 (t, 3, J ) 7.1); 13C NMR (partial) 172.2,
126.5, 125.7, 61.3, 59.1, 45.9, 34.6, 32.6, 31.4, 25.4, 14.1.
The data for 99a were determined from the mixture: 1H
NMR 5.90 (dddd, 1, J ) 11.0, 8.1, 2.0, 0.8), 5.81 (ddd, 1, J )
11.0, 6.4, 4.6), 4.20 (q, 2, J ) 7.1), 2.98 (br dt, 1, J ) 2.0, 8.1),
2.62 (dddd, 1, J ) 13.2, 10.7, 4.0, 0.8), 2.53-2.21 (m, 4), 2.10
(ddd, 1, J ) 13.3, 9.3, 5.9), 1.98-1.87 (m, 2), 1.27 (t, 3, J )
7.1); 13C NMR (partial) 62.5, 49.3, 41.9, 35.5, 34.3, 21.2, 14.0;
IR (neat) 1732, 1716.
The data for 86: 1H NMR 7.80 (dd, 1, J ) 7.8, 1.2), 7.51
(ddd, 1, J ) 7.6, 7.6, 1.5), 7.38-7.25 (m, 2), 3.69 (s, 1, OH),
2.66 (ddd, 1, J ) 8.4, 6.0, 2.4), 2.51 (ddd, 1, J ) 9.5, 9.5, 2.4),
2.25-1.23 (m, 8), 0.77 (t, 3, J ) 7.3); 13C NMR 203.6, 146.5,
139.4, 133.3, 129.5, 127.3, 127.1, 83.4, 49.8, 44.9, 42.9, 34.8,
32.2, 24.2, 11.5.
The data for 87: 1H NMR (partial data) 7.30-7.20 (m, 3),
7.15 (dd, 1, J ) 6.9, 1.7), 4.00 (s, 1, OH), 3.13 (dd, 1, J ) 8.7,
3.7), 2.95 (m, 1), 0.89 (t, 3, J ) 7.3); 13C NMR 200.1, 140.8,
127.6, 127.5, 126.9, 126.8, 125.4, 77.8, 51.6, 49.9, 45.8, 36.2,
32.5, 20.0, 11.5.
IR of the mixture: (CCl4) 3521, 1724, 1687.
Hyd r olysis of 83b To Give 86 a n d 87. To 5 mL of 1 N
NaOH solution was added a solution of 83b (20 mg, 0.073
mmol) in 5 mL of ethanol. The resulting mixture was stirred
at rt for 3 h. The mixture was neutralized by addition of 5
mL of 1 N HCl solution and extracted with CH2Cl2 (3 × 5 mL).
The combined organic layers were washed with brine, dried
(Na2SO4), and concentrated in vacuo to yield 14 mg (83%) of a
1:1 mixture of 86 and 87.
Oxid a tive Cycliza tion of 2-Meth yl-2-(2-p r op en yl)-1,3-
cyclop en ta n ed ion e (89a ). Reaction of 89a (76 mg, 0.5
mmol), Mn(OAc)3‚2H2O (303 mg, 1.125 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 2.5 mL of degassed AcOH
at 80 °C for 2 h followed by normal workup gave 79 mg of a
pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 3 mg (4%) of a 2:1 mixture of 90a and
92a , followed by 52 mg (70%) of 90a , 5 mg (6%) of a 1:1.5
mixture of 90a and 91a , and 4 mg (5%) of 91a . The calculated
yields are 75% of 90a , 9% of 91a , and 2% of 92a .
The data for 100a : 1H NMR 4.75 (t, 1, J ) 1.5), 4.71 (br s,
1), 4.22 (q, 2, J ) 7.1), 3.12 (d, 1, J ) 6.6), 2.67-2.55 (m, 2),
2.36 (ddd, 1, J ) 14.4, 5.8, 1.3), 2.25-2.19 (m, 4), 1.88-1.80
(m, 1), 1.28 (t, 3, J ) 7.1); 13C NMR 210.1, 171.2, 148.9, 109.8,
61.2, 57.2, 55.9, 35.6, 26.9, 26.7, 22.7, 14.1; IR (neat) 1742,
1712.
The data for 101a : 1H NMR 4.95-5.05 (br s, 2), 5.01 (dd,
1, J ) 6, 6), 4.25 (q, 2, J ) 7.1), 2.70 (dddd, 1, J ) 13.9, 10.1,
2.7, 1.3), 2.53-2.38 (m, 3), 2.33 (ddd, 1, J ) 13.9, 11.9, 6.0),
2.06 (ddd, 1, J ) 13.8, 10.8, 6.2), 2.07-1.97 (m, 1), 1.91 (dddd,
1, J ) 13.8, 10.8, 2.8, 1.3), 1.30 (t, 3, J ) 7.1); 13C NMR (partial)
171.6, 170.9, 144.6, 115.1, 80.7, 61.9, 31.9, 29.0, 25.5, 24.7, 14.0;
IR (neat) 1732.
The data for 90a : 1H NMR 6.05 (dddd, 1, J ) 9.2, 6.6, 2.4,
1.1), 5.62 (ddd, 1, J ) 9.2, 3.0, 3.0), 3.15 (t, 1, J ) 6.6), 2.97 (d,
1, J ) 17.8), 2.80-2.61 (m, 3), 1.15 (s, 3); 13C NMR 213.8, 211.7,
133.4, 126.4, 57.4, 50.2, 47.7, 45.4, 12.6; IR (neat) 1769, 1732.
The data are identical to those previously reported.26
The data for 91a : 1H NMR 5.81 (dt, 1, J ) 8.8, 2.7), 5.27
(dd, 1, J ) 8.8, 2.9), 3.08 (br d, 1, J ) 17.5), 2.85-2.68 (m, 4),
1.24 (s, 3).
Oxid a tion of â-Keto Ester 100a to La cton e 101a .
A
1
The H NMR data for 92a were obtained from the mixture:
7.23 (s, 2), 5.57-5.42 (m, 1), 5.08 (br d, 1, J ) 10.7), 5.06 (br
d, 1, J ) 16.6), 2.28 (d, 2, J ) 7.4), 1.15 (s, 3).
solution of 100a (26 mg, 0.125 mmol) and Mn(OAc)3‚2H2O (67
mg, 0.25 mmol) in 2.5 mL of degassed AcOH was stirred at 80
°C for 14 h until the solution turned bright blue, which

7846 J . Org. Chem., Vol. 61, No. 22, 1996
Cole et al.
indicated that all of the Mn(III) was consumed. Normal
workup afforded 23 mg of a 1:1 crude mixture of 100a and
101a .
mg, 0.42 mmol) were stirred in 10 mL of glacial acetic acid at
85 °C for 15 h. Normal workup and flash chromatography of
the residue on silica gel (30:1 hexane/EtOAc) recovered 12 mg
(11%) of unreacted 102b, followed by 70 mg (75%) of a 5:2
mixture of 103b and 104b. 103b: 1H NMR 5.48 (br dd, 1, J
) 13.5, 5.8), 5.39 (m, 1), 4.21 (q, 2, J ) 7.1), 2.70-2.36 (m, 4),
2.18-1.84 (m, 6), 1.63 (d, 3, J ) 5.8), 1.62 (m, 1), 1.40 (m, 1),
1.28 (t, 3, J ) 7.1); 13C NMR 214.6, 172.7, 131.8, 126.0, 61.1,
57.9, 51.2, 46.4, 35.8, 35.2, 27.7, 26.6, 20.7, 18.1, 14.1. 104b:
1H NMR (partial data) 4.22 (q, 2, J ) 7.1), 1.70 (d, 3, J ) 5.8),
1.29 (t, 3, J ) 7.1); 13C NMR 214.6, 173.0, 130.7, 125.5, 61.1,
57.7, 50.8, 41.3, 36.0, 35.4, 27.9, 26.6, 20.8, 14.1, 13.1; IR (neat)
1734, 1715.
Oxid a tive Cycliza tion of Meth yl 1-(3(Z)-Hexen yl)-3-
m eth yl-2-oxocyclop en ta n eca r boxyla te (102c). Mn(OAc)3‚
2H2O (107 mg, 0.40 mmol), Cu(OAc)2 (29 mg, 0.16 mmol), and
102c (38 mg, 0.16 mmol) were stirred in 3.5 mL of glacial acetic
acid at 70 °C for 12 h. Normal workup and flash chromatog-
raphy of the residue on silica gel (30:1 hexane/EtOAc) provided
27 mg (72%) of an 8:1 mixture of 103c and 104c. 103c: 1H
NMR 5.47 (dq, 1, J ) 15.1, 6.4), 5.30 (ddq, 1, J ) 15.1, 8.3,
1.4), 3.75 (s, 3), 2.50 (dddd, 1, J ) 12.8, 12.8, 4.2, 1.5), 2.24
(m, 1), 2.17 (br dd, 1, J ) 12.6, 7.3), 2.10-2.00 (m, 2), 1.90
(ddd, 1, J ) 12.8, 10.9, 4.7), 1.67 (dd, 3, J ) 6.4, 1.4), 1.72-
1.49 (m, 3), 0.92 (s, 3); decoupling at δ 5.30 collapsed the peak
at δ 2.24 (br dd, 1, J ) 10.0, 6.8), indicating that the H is axial
and the propenyl substituent is equatorial; 13C NMR 216.0,
172.2, 129.2, 127.7, 58.2 (q), 54.2, 52.3, 50.7 (q), 36.4, 26.4,
25.7, 25.6, 18.0, 17.8. 104c: 1H NMR (partial data) 3.76 (s,
3), 2.68 (m, 1), 0.88 (s, 3); 13C NMR (partial data) 128.7, 126.1,
48.1, 36.7, 17.3, 13.3; IR (neat) 1750, 1729.
Oxid a tive Cycliza tion of Eth yl 1-(3-Bu ten yl)-2-oxocy-
cloh exa n eca r boxyla te (97b). Reaction of â-keto ester 97b
(112 mg, 0.5 mmol), Mn(OAc)3‚2H2O (402 mg, 1.5 mmol), and
Cu(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed AcOH
at 80 °C for 18 h followed by normal workup gave 105 mg of
a pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 10 mg (9%) of a 2:3 mixture of
recovered 97b and 98b, followed by 22 mg (20%) of 100b, 28
mg (25%) of a 1:1 mixture of 100b and 99b, 33 mg (30%) of a
3:1 mixture of 99b and 100b, and 3 mg (3%) of a 6:1 mixture
of 99b and 100b. The calculated yields are 6% of 98b, 37% of
99b, and 40% of 100b.
1
The H NMR data for 98b were obtained from the mixture:
5.92 (dddd, 1, J ) 11.0, 7.8, 4.7, 1.4), 5.42 (ddd, 1, J ) 11.0,
5.5, 2.4), 4.23 (q, 2, J ) 7.2), 3.29-3.25 (m, 1), 2.48-2.26 (m,
3), 2.24-2.08 (m, 2), 2.04-1.82 (m, 3), 1.76 (ddd, 1, J ) 14.0,
2.9, 0.8), 1.68-1.58 (m, 1), 1.29 (t, 3, J ) 7.2).
1
The H NMR data for 99b were obtained from the mixture:
5.88-5.83 (m, 2), 4.22 (q, 2, J ) 7.1), 2.81-2.77 (m, 1), 2.60
(br d, 2, J ) 4.8), 2.51-2.38 (m, 2), 2.36-2.19 (m, 2), 2.13-
1.86 (m, 3), 1.62-1.55 (m, 1), 1.28 (t, 3, J ) 7.1). The data
are identical to those previously reported.26
The data for 100b: 1H NMR 4.92 (s, 1), 4.81 (s, 1), 4.22 (q,
2, J ) 7.1), 3.17 (br s, 1), 2.79-2.69 (m, 1), 2.60 (dddd, 1, J )
13.6, 8.4, 6.5, 1.5), 2.51-2.45 (m, 1), 2.34 (ddt, 1, J ) 14.9,
1.0, 7.2), 2.81-1.98 (m, 5), 1.67-1.61 (m, 1), 1.29 (t, 3, J )
7.1); 13C NMR 211.9, 172.5, 148.7, 111.2, 61.2, 57.8, 55.5, 37.6,
36.8, 32.7, 30.6, 18.9, 14.1; IR (neat) 1734, 1708.
Oxid a tive Cycliza tion of Eth yl 2-(3-Bu ten yl)-2-m eth yl-
3-oxob u t a n oa t e (105a ). Mn(OAc)3‚2H2O (516 mg, 1.92
mmol), Cu(OAc)2 (108 mg, 0.59 mmol), and 105a (117 mg, 0.59
mmol) were stirred in 5 mL of glacial acetic acid at 80 °C for
40 h. Normal workup and flash chromatography of the residue
on silica gel (30:1 hexane/EtOAc) gave 20 mg (17%) of unre-
acted 105a , followed by 1 mg (1%) of 107a , 13 mg (10%) of
110a , and 48 mg (32%) of 109a .
Oxid a tive Cycliza tion of Eth yl 2-Oxo-1-(4-p en ten yl)-
cyclop en ta n eca r boxyla te (97c). Reaction of 97c (112 mg,
0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed AcOH at
80 °C for 2.5 h followed by normal workup gave 104 mg of a
pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 4 mg (4%) of a 1:1 mixture of 100c
and 99c, followed by 5 mg (5%) of 99c, 4 mg (4%) of a 3:1
mixture of 99c and 98c, and 81 mg of polymeric material. The
calculated yields are 1% of 98c, 10% of 99c, and 2% of 100c.
Partial 1H NMR data for 98c were obtained from the
mixture: 5.91-5.84 (m, 2), 4.22 (q, 2, J ) 7.2), 2.84-2.75 (m,
1), 2.60 (m, 1), 1.25 (t, 3, J ) 7.2).
The data for 107a : 1H NMR 5.75 (m, 2), 4.19 (q, 2, J ) 7.1),
2.92 (m, 1), 2.63 (ddd, 1, J ) 13.3, 8.0, 5.3), 2.45 (m, 1), 2.23
(m, 2), 1.96 (m, 1), 1.38 (s, 3), 1.25 (t, 3, J ) 7.1).
The data for 109a : 1H NMR 6.05 (s, 1), 4.66 (s, 2), 4.17 (q,
2, J ) 7.1), 2.53 (m, 1), 2.39 (m, 1), 2.26 (ddd, 1, J ) 19.0, 5.0,
4.7), 2.13 (s, 3), 1.91 (ddd, 1, J ) 13.5, 8.6, 5.0), 1.39 (s, 3),
1.23 (t, 3, J ) 7.1); 13C NMR 196.3, 172.4, 170.1, 157.0, 123.7,
64.7, 61.4, 52.9, 32.9, 24.0, 20.6, 20.3, 14.0; IR (neat) 1740,
1732, 1682.
The data for 99c: 1H NMR 5.83 (ddd, 1, J ) 10.5, 7.4, 7.0),
5.76 (ddd, 1, J ) 10.5, 7.4, 7.0), 4.14 (q, 2, J ) 7.1), 2.68 (dddd,
1, J ) 7.3, 7.3, 7.1, 0.9), 2.39-2.28 (m, 4), 2.27-2.01 (m, 3),
1.83 (ddt, 1, J ) 12.9, 1.1, 4.0), 1.55-1.46 (ddd, 1, J ) 14.6,
11.5, 4.4), 1.31-1.22 (m, 1), 1.22 (t, 3, J ) 7.1); 13C NMR 194.5,
173.1, 132.4, 129.1, 61.3, 61.1, 51.3, 32.8, 27.8, 27.7, 26.4, 23.6,
14.0; IR (neat) 1747, 1728.
The data for 110a : 1H NMR 9.77 (s, 1), 6.62 (br s, 1), 4.17
(q, 2, J ) 7.1), 2.62-2.38 (m, 3), 1.91 (ddd, 1, J ) 13.6, 8.6,
5.0), 1.43 (s, 3), 1.23 (t, 3, J ) 7.1); 13C NMR 204.0, 193.8,
171.9, 153.1, 138.5, 61.7, 54.0, 32.8, 20.1, 19.6, 14.0; IR (neat)
1728, 1694, 1680.
Oxid a tive Cycliza tion of Eth yl 2-Meth yl-3-oxo-2-(4-
p en ten yl)bu ta n oa te (105b). Mn(OAc)3‚2H2O (643 mg, 2.40
mmol), Cu(OAc)2 (109 mg, 0.60 mmol), and 105b (127 mg, 0.60
mmol) were stirred in 25 mL of glacial acetic acid at 80 °C for
40 h. Normal workup and flash chromatography of the residue
on silica gel (20:1 hexane/EtOAc) gave 12 mg of 107b, followed
by 29 mg of a 2:3 mixture of 107b and unreacted 105b, 8 mg
of 106b, 9 mg of a 3:2 mixture of 106b and 109b, and 3 mg of
109b. The calculated yields of 107b, 106b, and 109b are 19%,
8%, and 4%, respectively.
Partial 1H NMR data for 100c were obtained from the
mixture: 4.88 (br s, 1), 4.81 (d, 1, J ) 1.0), 4.23 (q, 2, J ) 7.1),
3.28-3.20 (m, 1), 3.08-3.01 (m, 1), 1.23 (t, 3, J ) 7.1).
Oxid a tive Cycliza tion of Meth yl 1-(3(Z)-Hexen yl)-2-
oxocyclop en ta n eca r boxyla te (102a ). Reaction of 102a (116
mg, 0.52 mmol), Mn(OAc)3‚2H2O (416 mg, 1.55 mmol), and Cu-
(OAc)2 (94 mg, 0.52 mmol) in 10 mL of AcOH at 80 °C for 25
h followed by normal workup provided 122 mg of crude
product, which was purified by flash chromatography on silica
gel (30:1 hexane/EtOAc) to yield 83 mg (72%) of a 5:2 mixture
of 103a and 104a . 103a : 1H NMR 5.47 (ddq, 1, J ) 15.4, 6.2,
1.2), 5.32 (ddq, 1, J ) 15.4, 7.5, 1.2), 3.75 (s, 3), 2.70 (br ddd,
1, J ) 10.6, 7.5, 6.3), 2.55 (ddd, 1, J ) 12.5, 12.2, 4.2), 2.34 (br
d, 1, J ) 5.7), 2.20 (ddd, 1, J ) 12.6, 7.0, 1.2), 2.08-1.79 (m,
4), 1.74-1.58 (m, 2), 1.66 (dd, 3, J ) 6.2, 1.2); decoupling at δ
5.32 collapsed the peak at δ 2.69 (br dd, 1, J ) 10.6, 6.3); 13C
NMR 214.4, 171.9, 131.4, 125.7, 57.3, 52.2, 50.8, 48.2, 36.4,
27.2, 23.5, 17.9, 17.5. 104a : 1H NMR (partial data) 3.76 (s,
3), 3.05 (dddd, 1, J ) 10, 8, 6, 2.3); decoupling at δ 5.32
collapsed the peak at δ 3.04 (ddd, 1, J ) 10, 6, 2.3); 13C NMR
214.1, 171.9, 130.4, 125.1, 57.3, 52.2, 50.8, 43.3, 36.8, 27.3, 24.4,
17.5, 13.1; IR (neat) 1756, 1729.
The data for 106b: 1H NMR 6.04 (dd, 1, J ) 13.2, 1.5), 5.97
(dd, 1, J ) 13.2, 2.9), 5.25 (m, 1), 4.15 (q, 2, J ) 7.1), 2.36 (m,
1), 2.03 (s, 3), 2.22-1.92 (m, 2), 1.70 (m, 2), 1.38 (s, 3), 1.25
(m, 1), 1.23 (t, 3, J ) 7.1); 13C NMR 190.1, 172.4, 170.1, 133.4,
129.0, 72.0, 61.6, 58.0, 32.1, 27.8, 21.1, 20.2, 17.8, 14.0; IR
(neat) 1738, 1732, 1698.
The data for 107b: 1H NMR 5.62 (m, 2), 4.25 (q, 2, J ) 7.1),
3.13 (m, 1), 2.58-2.40 (m, 2), 2.28-2.03 (m, 3), 1.37 (s, 3),
1.43-1.21 (m, 2), 1.30 (t, 3, J ) 7.1); 13C NMR 210.3, 174.2,
130.3, 129.6, 61.1, 57.7, 43.5, 35.2, 24.5, 23.2, 22.6, 14.1; IR
(neat) 1734, 1700.
Oxid a tive Cycliza tion of Eth yl 1-(3(Z)-Hexen yl)-2-
oxocycloh exa n eca r boxyla te (102b). Mn(OAc)3‚2H2O (284
mg, 1.06 mmol), Cu(OAc)2 (77 mg, 0.42 mmol), and 102b (107
The data for 109b: 1H NMR 6.08 (s, 1), 4.55 (s, 2), 4.14 (q,
2, J ) 7.1), 2.50-1.94 (m, 2), 2.12 (s, 3), 1.79-1.45 (m, 2), 1.40

Free-Radical Cyclizations of Unsaturated Ketones
J . Org. Chem., Vol. 61, No. 22, 1996 7847
(s, 3), 1.34-1.20 (m, 2), 1.21 (t, 3, J ) 7.1); 13C NMR 199.9,
173.6, 172.4, 149.3, 127.0, 66.6, 61.3, 58.2, 34.1, 31.1, 23.4, 23.1,
20.8, 14.0; IR (neat) 1738, 1735, 1685.
The data for 123: 1H NMR 5.80 (ddd, 1, J ) 17.3, 10.5, 5.9),
5.19 (d, 1, J ) 17.3), 5.16 (d, 1, J ) 10.5), 4.19 (q, 2, J ) 7.1),
2.88 (br d, 1, J ) 5.9), 2.77 (br d, 1, J ) 4.1), 2.63 (br d, 1, J
) 5.1), 2.06 (dddd, 1, J ) 13.6, 11.3, 5.1, 4.8), 1.76 (dddd, 1, J
) 13.4, 11.3, 5.2, 4.1), 1.70 (br ddd, 1, J ) 13.4, 8.1, 4.8), 1.48
(br ddd, 1, J ) 13.6, 8.1, 5.2), 1.33 (s, 3), 1.27 (t, 3, J ) 7.1);
13C NMR 213.2, 171.7, 133.9, 117.5, 61.5, 61.1, 53.3, 49.1, 48.7,
23.6, 19.5, 17.7, 14.1; IR (neat) 1752, 1728. Anal. Calcd for
C13H18O3: C, 70.24; H, 8.16. Found: C, 69.98; H, 8.26.
Partial data for 122 were determined from the mixture. 1H
NMR 6.95 (ddd, 1, J ) 10.2, 5.8, 1.7), 6.09 (br d, 1, J ) 10.2),
5.58-5.55 (m, 1), 1.73 (br d, 3, J ) 6.8), 1.35 (s, 3), 1.22 (t, 3,
J ) 7.1).
Oxid a tive Cycliza tion of Eth yl 2-(3(Z)-Hexen yl)-2-
m eth yl-3-oxobu ta n oa te (111). Mn(OAc)3‚2H2O (422 mg,
1.57 mmol), Cu(OAc)2 (72 mg, 0.39 mmol), and 111 (89 mg,
0.39 mmol) were stirred in 6 mL of glacial acetic acid at 80 °C
for 40 h. Normal workup and flash chromatography of the
residue on silica gel (30:1 hexane/EtOAc) gave 12 mg (14%) of
112 containing 5-10% of the Z-isomer, followed by 25 mg
(28%) of unreacted 111 and 11 mg (10%) of 113 containing
5-10% of the Z-isomer.
The data for 112: (E-isomer) 1H NMR 5.43 (dq, 1, J ) 15.3,
5.6), 5.34 (br dd, 1, J ) 15.3, 4.9), 4.18 (q, 2, J ) 7.1), 2.54-
2.30 (m, 4), 1.76 (m, 1), 1.65 (d, 3, J ) 5.6), 1.62-1.41 (m, 2),
1.27 (s, 3), 1.26 (t, 3, J ) 7.1); 13C NMR 207.3, 172.9, 134.1,
124.3, 61.3, 56.2, 46.9, 42.7, 36.8, 29.5, 21.1, 17.8, 14.1; (Z-
isomer) 1H NMR (partial data) 5.20 (ddq, 1, J ) 10.8, 9.1, 1.7);
IR (neat) 1739, 1715.
Oxid a tive Cycliza tion of 120 a n d 121. Reaction of 30
mg (0.15 mmol) of a 1:2 mixture of 120 and 121, respectively,
Mn(OAc)3‚2H2O (100 mg, 0.375 mmol), and Cu(OAc)2‚H2O (30
mg, 0.15 mmol) in 5 mL of degassed AcOH at 80 °C for 16 h
followed by normal workup gave 26 mg of a pale yellow oil.
Purification by flash chromatography (10:1 hexane/EtOAc)
afforded 21 mg (70%) of a 1:2 mixture of 124 and 125.
Partial data for 124 were determined from the mixture: 1H
NMR 5.77 (ddd, 1, J ) 17.2, 10.3, 6.2), 5.20 (ddd, 1, J ) 17.2,
1.5, 1.5), 5.12 (ddd, 1, J ) 10.3, 1.5, 1.5), 4.21 (q, 2, J ) 7.1),
2.76 (br d, 1, J ) 6.2), 2.63 (br d, 1, J ) 4.9), 2.47 (br d, 1, J
) 3.7), 1.41 (s, 3), 1.27 (t, 3, J ) 7.1).
The data for 113: (E-isomer) 1H NMR 5.50 (dq, 1, J ) 15.3,
6.4), 5.29 (ddq, 1, J ) 15.3, 7.5, 1.6), 5.16 (d, 1, J ) 11.9), 4.29
(dq, 1, J ) 10.7, 7.1), 4.18 (dq, 1, J ) 10.7, 7.1), 2.49 (ddd, 1,
J ) 13.9, 3.3, 3.3), 2.44 (m, 1), 2.15 (s, 3), 1.80 (m, 1), 1.65 (dd,
3, J ) 6.4, 1.6), 1.49-1.21 (m, 2), 1.32 (s, 3), 1.30 (t, 3, J )
7.1); 13C NMR 200.9, 172.1, 170.0, 130.5, 127.3, 78.7, 61.9, 56.7,
47.5, 36.1, 27.6, 20.9, 20.6, 17.9, 13.9; (Z-isomer) 1H NMR
(partial data) 2.12 (s, 3), 1.34 (s, 3); IR (neat) 1755, 1744, 1724.
Oxid a tive Cycliza tion of Eth yl 2-Oxo-1-(p h en ylm eth -
yl)cyclop en ta n eca r boxyla te (114). Reaction of 114 (123
mg, 0.5 mmol), Mn(OAc)3‚2H2O (335 mg, 1.25 mmol), and Cu-
(OAc)2‚H2O (100 mg, 0.5 mmol) in 5 mL of degassed AcOH at
80 °C for 2 h followed by normal workup gave 136 mg of a
pale yellow oil. Purification by flash chromatography (20:1
hexane/EtOAc) afforded 36 mg (30%) of a 3:1 mixture of 115
and recovered 114, followed by 12 mg (10%) of 115, 6 mg (5%)
of a 2:1 mixture of 115 and ethyl 2-oxo-1-(phenylmethyl)-
cyclopent-3-enecarboxylate (enone), 9 mg (7%) of enone, 3 mg
(2%) of a 1:1 mixture of enone and ethyl 3-acetoxy-2-oxo-1-
(phenylmethyl)cyclopentanecarboxylate (acetates), and 45 mg
(37%) of a 1:1 mixture of acetates. The calculated yields are
35% of 115, 10% of enone, and 39% of acetates.
Partial data for 125 were determined from the mixture. 1H
NMR 6.98-6.89 (m, 1), 6.11 (ddd, 1, J ) 10.2, 2.0, 1.6), 5.48
(br t, 1, J ) 6.9).
Oxid a tive Cycliza tion of Meth yl 1-(2(Z),6(Z)-Non d i-
en yl)-2-oxocyclop en t a n eca r b oxyla t e (126). Mn(OAc)3‚
2H2O (160 mg, 0.60 mmol), Cu(OAc)2 (48 mg, 0.26 mmol), and
126 (70 mg, 0.26 mmol) were stirred in 7 mL of glacial acetic
acid at 80 °C for 15 h. Normal workup and flash chromatog-
raphy of the residue on silica gel (30:1 hexane/EtOAc) provided
8 mg of unreacted 126, followed by 6 mg of a 3:2 mixture of
126 and 128, followed by 21 mg of a 1.3:4.3:1 mixture of 126,
128, and 129. The calculated yields of 128 and 129 are 23%
and 5%, respectively.
The data for 128: 1H NMR 5.54-5.31 (m, 2), 5.28 (m, 1),
3.76 (s, 3), 3.21 (dd, 1, J ) 17.0, 2.1), 2.66-2.49 (m, 4), 2.17-
1.74 (m, 8), 0.96 (t, 3, J ) 7.4); IR (neat) 1760, 1732. The NMR
spectral data are very similar to those of 8e.
The data for 115: 1H NMR 7.25-6.96 (m, 4), 4.27 (q, 2, J )
7.1), 3.89 (br d, 1, J ) 16.8), 3.36 (d, 1, J ) 16.8), 3.28 (d, 1, J
) 5.6), 2.74 (br ddd, 1, J ) 12.1, 10.3, 2.2), 2.34-2.25 (m, 1),
2.17-2.01 (m, 2), 1.32 (t, 3, J ) 7.1); 13C NMR (partial) 130.2,
128.6, 128.3, 127.5, 127.4, 127.1, 61.0, 55.8, 51.2, 45.3, 29.7,
29.4, 14.2; IR (neat) 1738, 1709.
Partial data for 129: 1H NMR 3.79 (s, 3), 0.95 (t, 3, J )
7.5).
Oxid a tive Cycliza tion of Meth yl 1-(3(Z)-Hexen yl)-2-
oxo-3-(2-p r op en yl)cyclop en ta n eca r boxyla te (131). Mn-
(OAc)3‚2H2O (100 mg, 0.37 mmol), Cu(OAc)2 (30 mg, 0.17
mmol), and 131 (44 mg, 0.17 mmol) were stirred in 3.5 mL of
glacial acetic acid at 70 °C for 12 h. Normal workup and flash
chromatography of the residue on silica gel (40:1 hexane/
EtOAc) provided 36 mg (82%) of a 9:1 mixture of 133 and 134.
133: 1H NMR 4.96 (m, 1), 4.83 (m, 1), 3.76 (s, 3), 2.85 (dddd,
1, J ) 16.4, 2.9, 2.7, 2.7), 2.52 (dddd, 1, J ) 12.8, 12.8, 4.5,
1.8), 2.33-2.22 (m, 2), 2.12-1.96 (m, 3), 1.92-1.81 (m, 3),
1.70-1.50 (m, 4), 0.89 (t, 3, J ) 7.5); 13C NMR 212.7, 172.1,
151.7, 106.8, 59.1, 57.8, 56.5, 52.3, 47.4, 38.6, 36.7, 26.2, 25.6,
23.9, 23.2, 10.4. 134: 1H NMR (partial data) 5.00 (m, 1), 4.85
(m, 1), 3.74 (s, 3), 2.67 (m, 1), 0.84 (t, 3, J ) 7.5); 13C NMR
(partial data) 151.0, 106.6, 55.8, 52.2, 44.9, 35.9, 35.2, 27.0,
24.9, 22.4, 19.0, 9.6; IR (neat) 1754, 1730.
The data for enone: 1H NMR 7.58 (dt, 1, J ) 5.7, 2.7), 7.26-
7.09 (m, 5), 6.07 (dt, 1, J ) 5.7, 2.2), 4.19 (q, 2, J ) 7.1), 3.27
(s, 2), 3.16 (br ddd, 1, J ) 19.3, 2.7, 2.2), 2.72 (br ddd, 1, J )
19.3, 2.7, 2.2), 1.25 (t, 3, J ) 7.1).
The data for acetates: 1H NMR 7.28-7.06 (m, 4), 5.30 (dd,
0.5 × 1, J ) 10.7, 8.5), 4.64 (dd, 0.5 × 1, J ) 11.7, 8.3), 4.19
(q, 0.5 × 2, J ) 7.1), 4.18 (q, 0.5 × 2, J ) 7.1), 3.29-3.12 (m,
2), 2.41 (dd, 0.5 × 1, J )5.9, 13.3), 2.37-2.16 (m, 2), 2.11 (s,
0.5 × 3), 2.10 (s, 0.5 × 3), 2.01-1.79 (m, 1.5), 1.29 (t, 0.5 × 3,
J ) 7.1), 1.25 (t, 0.5 × 3, J ) 7.1).
Oxid a tive Cycliza tion of Eth yl 3-Meth yl-2-oxo-7-d e-
cen oa te (118). A solution of â-keto ester 118 (46 mg, 0.2
mmol), Mn(OAc)3‚2H2O (214 mg, 0.8 mmol), and Cu(OAc)2‚H2O
(40 mg, 0.2 mmol) in 5 mL of degassed AcOH was stirred under
N2 for 56 h at 60 °C. Normal workup yielded 41 mg of a pale
yellow oil. Purification by flash chromatography (10:1 hexane/
EtOAc) gave 5 mg (10%) of recovered 118, followed by 5 mg
(10%) of 119, 2 mg (4%) of a 2:1 mixture of 121 and 120, and
20 mg (45%) of a 17:1:1.7:0.1 mixture of 123, 124, 122, and
125.
Ack n ow led gm en t. We thank the National Insti-
tutes of Health for financial support. We thank Andrew
Maioli, Christopher Knoell, Timothy McKee, and
Lioudmila Zaslavskaia for experimental assistance.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental details for preparation of the cyclization substrates;
1H and 13C NMR spectra of many compounds (123 pages). This
material is contained in libraries on microfiche, immediately
follows this article in the microfilm version of the journal, and
can be ordered from the ACS; see any current masthead page
for ordering information.
Oxid a tive Cycliza tion of Eth yl 1-Meth yl-2-oxo-6-(1(E)-
p r op en yl)cycloh exa n eca r boxyla te (119). Reaction of pure
119 (44 mg, 0.2 mmol), Mn(OAc)3‚2H2O (134 mg, 0.5 mmol),
and Cu(OAc)2‚H2O (40 mg, 0.2 mmol) in 5 mL of degassed
AcOH at 80 °C for 16 h followed by normal workup gave 47
mg of a pale yellow oil. Purification by flash chromatography
(10:1 hexane/EtOAc) afforded 36 mg (83%) of a 10:1 mixture
of 123 and 122.
J O961199U