Is an iodine atom almighty as a leaving group for Bu3SnH-mediated radical cyclization? The effect of a halogen atom on the 5-endo-trig radical cyclization of N-vinyl-α-halo amides

Article abstract of DOI:10.1021/jo020007u

The effect of a halogen atom as a leaving group on Bu₃SnH-mediated 5-endo-trig radical cyclization of N-(cyclohex-1-enyl) α-halo amides was examined. The cyclization of α-chloro amides occurred with a high degree of efficiency, whereas the corresponding α-iodo congeners gave only limited quantities of cyclization products. A detailed study revealed that these phenomena could be attributed to the initial conformations of α-halo amides. The cyclizing ability of α-iodo amides can be restored with Bu₃SnCl or Bu₃SnF as an additive. The cyclization of an α-iodo amide in the presence of Bu₃SnF could be applied to a short-step synthesis of lycoranes featuring sequential 5-endo-trig and 6-endo-trig radical cyclizations.

Full text of DOI:10.1021/jo020007u

Is a n Iod in e Atom Alm igh ty a s a Lea vin g Gr ou p for
Bu 3Sn H-Med ia ted Ra d ica l Cycliza tion ? Th e Effect of a Ha logen
Atom on th e 5-En d o-tr ig Ra d ica l Cycliza tion of N-Vin yl-r-h a lo
Am id es
Osamu Tamura, Hana Matsukida, Atsushi Toyao, Yoshifumi Takeda, and
Hiroyuki Ishibashi*
Faculty of Pharmaceutical Sciences, Kanazawa University, Takara-machi, Kanazawa 920-0934, J apan
isibasi@mail.p.kanazawa-u.ac.jp
Received J anuary 2, 2002
The effect of a halogen atom as a leaving group on Bu3SnH-mediated 5-endo-trig radical cyclization
of N-(cyclohex-1-enyl) R-halo amides was examined. The cyclization of R-chloro amides occurred
with a high degree of efficiency, whereas the corresponding R-iodo congeners gave only limited
quantities of cyclization products. A detailed study revealed that these phenomena could be
attributed to the initial conformations of R-halo amides. The cyclizing ability of R-iodo amides can
be restored with Bu3SnCl or Bu3SnF as an additive. The cyclization of an R-iodo amide in the
presence of Bu3SnF could be applied to a short-step synthesis of lycoranes featuring sequential
5-endo-trig and 6-endo-trig radical cyclizations.
SCHEME 1
In tr od u ction
γ-Lactam formation using 5-exo or 5-endo cyclization
of a carbamoylmethyl radical has attracted considerable
attention over the past two decades.1,2 Since our report
on the Bu3SnH-mediated 5-endo-trig cyclization of N-(cy-
clohex-1-enyl)-N-alkyl-R-chloroacetamides (Scheme 1, X
) Cl),3a 5-endo cyclization of carbamoylmethyl radicals
of enamides has been extensively studied and applied to
syntheses of several natural products and related
compounds.3-6 This cyclization mode has therefore now
become a standard reaction for examining new methods
of generating carbamoylmethyl radicals.4-6
It is recognized that an iodine atom is a better leaving
group for Bu3SnH-mediated radical cyclizations of ω-halo
alkenes than is a bromine or chlorine atom7 because the
bond dissociation energy (BDE) of a carbon-iodine bond
is significantly lower than those of a carbon-bromine
* To whom correspondence should be addressed. Fax: +81(76)-234-
4476.
(1) For reviews, see: (a) Giese, B. Radicals in Organic Synthesis:
Formation of Carbon-Carbon Bonds; Pergamon: New York, 1986. (b)
Fossey, J .; Lefort, D.; Sorba, J . Free Radicals in Organic Chemistry;
J ohn Wiley & Sons: New York, 1995. (c) Curran, D. P. Synthesis 1988,
417, 489. (d) J asperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev.
1991, 91, 1237.
(2) For reviews on synthesis of heterocycles by radical cyclization,
see: (a) Bowman, W. R.; Bridge, C. F.; Brookes, P. J . Chem. Soc., Perkin
Trans. 1 2000, 1. (b) Bowman, W. R.; Cloonan, M. O.; Krintel, S. L. J .
Chem. Soc., Perkin Trans. 1 2001, 2885. For a review on 5-endo radical
cyclizations, see: (c) Ishibashi, H.; Sato, T.; Ikeda, M. Synthesis 2002,
695.
(3) For Bu3SnH-mediated 5-endo radical cyclization of N-vinyl-R-
halo amides, see: (a) Ishibashi, H.; Nakamura, N.; Sato, T.; Takeuchi,
M.; Ikeda, M. Tetrahedron Lett. 1991, 32, 1725. (b) Sato, T.; Nakamura,
N.; Ikeda, K.; Okada, M.; Ishibashi, H.; Ikeda, M. J . Chem. Soc., Perkin
Trans. 1 1992, 2399. (c) Sato, T.; Machigashira, N.; Ishibashi, H.; Ikeda,
M. Heterocycles 1992, 33, 139. (d) Sato, T.; Chono, N.; Ishibashi, H.;
Ikeda, M. J . Chem. Soc., Perkin Trans. 1 1995, 1115. (e) Ishibashi, H.;
Fuke, Y.; Yamashita, T.; Ikeda, M. Tetrahedron: Asymmetry 1996, 7,
2531. (f) Ishibashi, H.; Higuchi, M.; Ohba, M.; Ikeda, M. Tetrahedron
Lett. 1998, 39, 75. (g) Ikeda, M.; Ohtani, S.; Sato, T.; Ishibashi, H.
Synthesis 1998, 1803. (h) Ikeda, M.; Ohtani, S.; Yamamoto, T.; Sato,
T.; Ishibashi, H. J . Chem. Soc., Perkin Trans. 1 1998, 1763. (i) Ikeda,
M.; Hamada, M.; Yamashita, T.; Matsui, K.; Sato, T.; Ishibashi, H. J .
Chem. Soc., Perkin Trans. 1 1999, 1949. (j) Goodall, K.; Parsons, A. F.
J . Chem. Soc., Perkin Trans. 1 1994, 3257. (k) Goodall, K.; Parsons,
A. F. Tetrahedron 1996, 52, 6739. (l) Goodall, K.; Parsons, A. F.
Tetrahedron Lett. 1997, 38, 491. (m) Baker, S. R.; Parsons, A. F.;
Wilson, M. Tetrahedron Lett. 1998, 39, 2815. (n) Baker, S. R.; Parsons,
A. F.; Pons, J .-F.; Wilson, M. Tetrahedron Lett. 1998, 39, 7197. (o)
Baker, S. R.; Burton, K. I.; Parsons, A. F.; Pons, J .-F.; Wilson, M. J .
Chem. Soc., Perkin Trans. 1 1999, 427.
(4) For nickel-mediated 5-endo radical cyclization of N-vinyl-R-halo
amides, see: (a) Quiclet-Sire, B.; Saunier, J .-B.; Zard, S. Z. Tetrahedron
Lett. 1996, 37, 1397. (b) Cassayre, J .; Quiclet-Sire, B.; Saunier, J .-B.;
Zard, S. Z. Tetrahedron 1998, 54, 1029. (c) Cassayre, J .; Quiclet-Sire,
B.; Saunier, J .-B.; Zard, S. Z. Tetrahedron Lett. 1998, 39, 8995. (d)
Cassayre, J .; Zard, S. Z. Synlett 1999, 501. (e) Cassayre, J .; Dauge,
D.; Zard, S. Z. Synlett 2000, 471.
(5) For Cu(I)-mediated 5-endo radical cyclization of N-vinyl-R-halo
amides, see: (a) Davies, D. T.; Kapur, N.; Parsons, A. F. Tetrahedron
Lett. 1999, 40, 8615. (b) Clark, A. J .; Dell, C. P.; Ellard, J . M.; Hunt,
N. A.; McDonagh, J . P. Tetrahedron Lett. 1999, 40, 8619. (c) Clark, A.
J .; Filik, R. P.; Haddleton, D. M.; Radigue, A.; Sanders, C. J .; Thomas,
G. H.; Smith, M. E. J . Org. Chem. 1999, 64, 8954. (d) Bryans, J . S.;
Chessum, N. E. A.; Parsons, A. F.; Ghelfi, F. Tetrahedron Lett. 2001,
42, 2901.
(6) For Mn(III)-mediated 5-endo radical cyclization of N-vinyl-â-keto
or R-methylthio amides, see: (a) Davies, D. T.; Kapur, N.; Parsons, A.
F. Tetrahedron Lett. 1998, 39, 4397. (b) Davies, D. T.; Kapur, N.;
Parsons, A. F. Tetrahedron 2000, 56, 3941. (c) Ishibashi, H.; Toyao,
A.; Takeda, Y. Synlett 1999, 1468. (d) Toyao, A.; Chikaoka, S.; Takeda,
Y.; Tamura, O.; Muraoka, O.; Tanabe, G.; Ishibashi, H. Tetrahedron
Lett. 2001, 42, 1729.
(7) See ref 1b, pp 243-255.
10.1021/jo020007u CCC: $22.00 © 2002 American Chemical Society
Published on Web 07/09/2002
J . Org. Chem. 2002, 67, 5537-5545
5537

Tamura et al.
TABLE 1. Ra d ica l Cycliza tion of r-Ha lo Am id es 7a -c
SCHEME 2
SCHEME 3
products (% yield)
entry amide 7 conditionsa
2
8
9
2 + 8 + 9 10
1
2
3
4
5
7a
7b
7c
7a
7c
A
A
A
B
B
92
92
77
24
68
55 11 11
13 11
56
68
76
12
bond and a carbon-chlorine bond (BDE: CH3-I, 57 kcal/
mol; CH3-Br, 71 kcal/mol; CH3-Cl, 83 kcal/mol).8 There-
fore, the use of a chlorine atom as a leaving group for
the radical cyclizations of ω-halo alkenes frequently
resulted in an increase in the amount of uncyclized
reduction products due to an attack of the accumulated
Bu3SnH on the initially generated radicals. This was also
the case for 5-exo radical cyclization of N-(cyclohex-2-
enyl)-R-haloacetamides 1a ,b. Thus, Bu3SnH-mediated
reaction of R-iodo amide 1b gave a higher yield of
cyclization product 2 with a lower amount of simple
reduction product 3 than did that of R-chloro amide 1a
(Scheme 2).9
Recently, we reinvestigated Bu3SnH-mediated 5-endo
radical cyclization of N-(cyclohex-1-enyl)-R-haloaceta-
mides (Scheme 1; X ) Cl, Br, and I) and found that, in
contrast to the 5-exo cyclization described above, R-chloro
amides gave much higher yields of the 5-endo cyclization
products than did the corresponding R-iodo congeners and
that the cyclization abilities of the iodo congeners could
be restored by the use of Bu3SnCl as an additive.10 We
now give a full account of this work and its application
to a short-step synthesis of lycoranes using sequential
radical cyclizations.
a
Condition A: Bu3SnH (1.2 equiv), AIBN (0.1 equiv), toluene,
reflux. Condition B: (TMS)3SiH (1.2 equiv), AIBN (0.1 equiv),
toluene, reflux.
We further examined the effect of a halogen atom in
5-endo radical cyclization of R-halo amides by employing
N-benzyl amides 7a -c in place of N-methyl amides 4a ,b
(Table 1). The cyclization ability, interestingly, decreased
in the order of chloro amide 7a , bromo amide 7b, and
iodo amide 7c. Thus, treatment of 7a with Bu3SnH in
the presence of AIBN in boiling toluene gave cyclization
product 2 in 92% yield (entry 1). A similar treatment of
7b gave 2, enamide 8, and R,â-unsaturated product 9 in
55, 11, and 11% yields, respectively (entry 2). In contrast,
reaction of R-iodo amide 7c gave simple reduction product
10 in 68% yield accompanied by small amounts of
cyclization products 8 and 9 (entry 3). This tendency was
also observed in (TMS)3SiH-mediated reactions. Reaction
of R-chloro amide 7a with (TMS)3SiH in the presence of
AIBN gave 2 and 9 in 68% combined yield (entry 4),
whereas R-iodo amide 7c gave only simple reduction
product 10 in 76% yield (entry 5).
Formation of the cyclization products 2, 8, and 9 from
R-halo amides 7a -c may be rationalized as depicted in
Scheme 4. Carbamoylmethyl radical 11 generated from
7 causes 5-endo-trig cyclization to give R-amidoyl radical
12, which undergoes an attack of Bu3SnH or (TMS)3SiH,
affording 2. On the other hand, a single electron transfer
(SET) from R-amidoyl radical 12 to the starting R-halo
amides 7 generates cation 13 and an anion radical of 7,11
which releases a halide anion, giving rise to the required
radical 11. The cation 13 affords amides 8 and 9.12
Another possibility for the formation of 8 and 9 involves
an atom transfer reaction.13 Halogen atom transfer takes
place from R-halo amide 7 to radical 12 to give radical
Resu lts a n d Discu ssion
1. Effect of a Ha logen Atom . Our studies were begun
by comparing the effect of a halogen atom on the radical
cyclization of N-(cyclohex-1-enyl)-N-methyl-R-haloaceta-
mides 4a ,b (Scheme 3). In our previous study, on treat-
ment of R-chloro amide 4a with Bu3SnH in the presence
of AIBN in boiling toluene, 5-endo-trig radical cyclization
product 5 and simple reduction product 6 were obtained
in 63 and 8% yields, respectively.3a,b Treatment of the
corresponding iodo congener 4b under similar conditions,
surprisingly, gave only simple reduction product 6 with-
out any cyclization product 5.
(11) For SET reaction from a long-lived radical to a starting halide,
see refs 4b and 5c. See also: Marco-Contelles, J .; Rodrı´guez-Ferna´ndez,
M. Tetrahedron Lett. 2000, 41, 381.
(8) See ref 1b, pp 113, 298.
(12) Amide 8 appears to be the kinetic product from cation 13, and
R,â-unsaturated amide 9 may be the thermodynamic product produced
by isomerization of 8, because amide 8 was selectively obtained in some
cases (Table 2, entries 3 and 4, vide infra). Of course, the intermediate
for the formation of 9 from 8 might be cation 13.
(13) For iodine atom transfer reaction, see: Miranda, L. D.; Cruz-
Almanza, R.; Pavo´n, M.; Romero, Y.; Muchowski, J . M. Tetrahedron
Lett. 2000, 41, 10181 and references therein.
(9) Bu3SnH-mediated 5-exo-trig radical cyclization of the N-methyl
congener of 1b was reported to give a ca. 1:2 mixture of the cyclization
product 5 and the reduction product; see: (a) J olly, R. S.; Livinghouse,
T. J . Am. Chem. Soc. 1988, 110, 7536. (b) Curran, D. P.; Tamine, J . J .
Org. Chem. 1991, 56, 2746.
(10) Ishibashi, H.; Matsukida, H.; Toyao, A.; Tamura, O.; Takeda,
Y. Synlett 2000, 1497.
5538 J . Org. Chem., Vol. 67, No. 16, 2002

Effect of Halogen on 5-Endo-trig Radical Cyclization
SCHEME 4
F IGURE 1. Rotamers of R-halo amide 7a and 7c due to the
amide bonds.
carbon bond is much shorter than an iodine-carbon
bond,16 and hence the internal steric interaction between
the chlorine atom and the benzyl group in anti-7a may
be more severe than that between an iodine atom and
benzyl group in anti-7c.17 Therefore, the population of
syn-7a would be larger than that of syn-7c (ClK > IK).18,19
The above consideration appears to be supported by
the previous result with the N-methyl-substituted R-chlo-
ropropionamide 15a , which has more severe steric in-
teraction between both methyl groups than acetamide
derivative 4a (Scheme 5). Reaction of 15a with Bu3SnH
11 and halide 14, which affords enamide 8 along with
release of hydrogen halide. The present cyclizations
involving either the SET or the halogen atom transfer
require stoichiometric amounts (not catalytic amounts)
of Bu3SnH or (TMS)3SiH since the cyclizations are
accompanied by loss of hydrogen halides, which consume
Bu3SnH or (TMS)3SiH.11
(15) It was reported that the benzyl-axial conformer of cis-1-benzyl-
4-methylcyclohexane is slightly predominant over the corresponding
benzyl-equatorial one at low temperatures (for example, at -71 °C)
and that the benzyl-equatorial conformer becomes the major conformer
at room temperature due to the contribution of the entropic factor;
see: J uaristi, E.; Labastida, V.; Antu´nez, S. J . Org. Chem. 1991, 56,
4802. This strongly suggests that a benzyl group can be regarded as a
bulkier substituent than a methyl group at temperatures greater than
room temperature. The present radical reactions were carried out in
boiling toluene, and hence the steric interaction of the benzyl group
would be important.
(16) Bond lengths of C-X of CH2X are as follows: C-Cl ) 1.78 Å,
C-Br ) 1.94 Å, and C-I ) 2.14 Å. Equatorial-axial free energy
differences (∆G°, kcal/mol) of chloro-, bromo-, and iodocyclohexanes
are 0.53, 0.48, and 0.47, respectively; see: Carey, F. A.; Sandberg, R.
J . Advanced Organic Chemistry, 3rd ed.; Plenum Publishing: New
York, 1990; Part A, pp 133-135.
(17) In sulfonyl radical-mediated cyclization of diallylbenzylamines,
the use of the BH3 as a Lewis acid gave better diastereoselectivity than
the use of AlMe3. This result was explained by a conformational
analysis of the transition state for the cyclization of the amine-Lewis
acid complex. The bond length of N-B is shorter than that of N-Al,
so BH3 can make a more rigid conformation, resulting in a higher
degree of diastereoselectivity, than AlMe3. See: Bertrand, M.-P.;
Gastaldi, S.; Nouguier, R. Tetrahedron 1998, 54, 12829.
(18) The rotamers anti-7 and syn-7 are not detected in the NMR
time scale. The 1H NMR spectrum of 7a exhibits the signals due to
the vinyl proton and the allylic proton at δ 4.99 and 1.60, respectively,
whereas that of 7c exhibits signals at δ 5.20 and 1.75. Since the
carbonyl group of an anti conformer of R-halo amide 7 is located closer
to the cyclohexene ring than that of the syn conformer, the vinyl proton
and the allylic proton of the anti conformer may resonate more
downfield due to the anisotropy effect from the carbonyl group than
those of the syn conformer. The chemical shifts of the protons of 7a
and 7c seem to support that 7c has a higher population of the anti
conformer than 7a .
The tendency for reactions of R-chloro amides 4a and
7a to give higher yields of cyclization products than those
obtained by reactions of R-iodo congeners 4b and 7c is
thought to be due to the amides themselves, since the
reactions of 7a and 7c under different conditions showed
the same tendency (Table 1; entries 1 and 3 and entries
4 and 5, respectively). One possible explanation for the
difference between the modes of cyclization of R-chloro
and R-iodo amides may be derived from consideration of
the rotamers of the amides (Figure 1).14 Two conformers
anti-7a and syn-7a are considered for 7a and anti-7c and
syn-7c for 7c. The carbamoylmethyl radical 11 from syn-
7a ,c can cyclize, whereas that from anti-7a ,c cannot.
Taking into account the fact that 7a ,c, having bulkier
N-benzyl groups, afforded higher yields of cyclization
products than 4a ,c, bearing N-methyl groups, one of the
I
factors to determine equilibrium constants ClK and K
may be steric interaction between an N-substituent and
a halomethyl group.15 It is well-known that a chlorine-
(14) It has been suggested that the geometry of an initial rotamer
of a radical precursor plays a crucial role in determining the course of
a radical reaction; see: (a) Stork, G.; Mah, R. Heterocycles 1989, 28,
723. (b) Sato, T.; Wada, Y.; Nishino, M.; Ishibashi, H.; Ikeda, M. J .
Chem. Soc., Perkin Trans. 1 1989, 879. (c) Curran, D. P.; DeMello, N.
C. J . Chem. Soc., Chem. Commun. 1993, 1314. (d) Curran, D. P.; Liu,
W.; Chen, C. H.-T. J . Am. Chem. Soc. 1999, 121, 11012. (e) Musa, O.
M.; Horner, J . H.; Newcomb, M. J . Org. Chem. 1999, 64, 1022. (f)
Miyata, O.; Nishiguchi, A.; Ninomiya, I.; Aoe, K.; Okamura, K.; Naito,
T. J . Org. Chem. 2000, 65, 6922. (g) Besev, M.; Engman, L. Org. Lett.
2000, 2, 1589. (h) Ishibashi, H.; Kato, I.; Takeda, Y.; Kogure, M.;
Tamura, O. Chem. Commun. 2000, 1527. (i) Iwamatsu, S.; Matsubara,
K.; Nagashima, H. J . Org. Chem. 1999, 64, 9625. (j) Nagashima, H.;
Isono, Y.; Iwamatsu, S. J . Org. Chem. 2001, 66, 315 and references
therein; See also ref 9b.
(19) Another assumption for the lesser cyclization ability of 4b and
7c may involve steric interaction between R-halomethyl groups and
cyclohexene rings in R-halo acetamides 4 and 7; that is, steric repulsion
between the iodomethyl group and the cyclohexene ring in syn-7c
might be larger than that between the chloromethyl group and the
cyclohexene ring in syn-7a , which may result in the larger population
of anti-7c than that of anti-7a . However, this assumption can be ruled
out by considering the fact that radical reaction of 15a gave 16a in
73% yield without a reduction product (Scheme 5, vide infra) because
if the hypothesis was correct, radical reaction of R-chloro propionamide
15a , having a more sterically demanding haloalkyl group, should give
a lower yield of cyclization product 16a with a higher yield of the
reduction product than that of 6 from R-chloroacetamide 4a .
J . Org. Chem, Vol. 67, No. 16, 2002 5539

Tamura et al.
SCHEME 5
TABLE 2. Ra d ica l Cycliza tion of 7c in th e P r esen ce of
Ad d itives
F IGURE 2. Plausible role of Bu3SnCl on the radical cycliza-
tion of 7c.
tions of complex 7c-Sn formed from 7c and Bu3SnCl as
a Lewis acid (Figure 2).21 The anti-7c-Sn might have
severe steric repulsion between the cyclohexene ring and
the tin group, and hence the equilibrium might therefore
shift to syn-7c-Sn (IK < Sn-IK). Another possibility may
be due to the relatively long lifetime of radical 18
generated from 7c-Sn. Radical anti-11 generated from
anti-7c undergoes facile attack of Bu3SnH to afford
simple reduction product 10, whereas the radical center
of anti-18 derived from anti-7c-Sn is efficiently blocked
from attack of Bu3SnH by sterically hindered Bu3SnCl;
thereby, radical anti-18 gains sufficient lifetime to isomer-
ize to syn-18, which is the favored conformation for
cyclization.22 The fact that Bu3SnI is much less efficient
as an additive than Bu3SnCl or Bu3SnF may be ascribed
to the lack of Lewis acidity.
Parsons and co-workers also reported that the cycliza-
tion abilities of N-(R-haloacetamido)dehydroalanines 19
decreased in the order of the use of Cl, Br, and I as
leaving groups (the yields of 20 for chloro amide, bromo
amide, and iodo amide are 52, 47, and 38%, respectively),
although they gave no explanation for the phenomenon.3k
Therefore, we reinvestigated Bu3SnH-mediated radical
cyclization of Parsons’s substrates 19a ,b (Table 3). Treat-
products (% yield)
entry
additive (equiv)
2
8
9
2 + 8 + 9
10
1
2
3
4
5
6
Bu3SnCl (1)
Bu3SnCl (5)
Bu3SnF (5)
Bu3SnOAc (5)
Bu3SnI (5)
11
18
9
13
20
54
52
14
14
28
38
66
63
52
26
24
30
9
12
18
71
48
(TMS)3SiCl (5)
6
in the presence of AIBN gave no reduction product and
afforded 5-endo-trig cyclization product 16a (a mixture
of diastereomers) in 73% yield (compared with the
reaction of 4a ).3a,b R-Iodo congener 15b also gave cycliza-
tion product 16b, although the yield was low (24%),
together with the reduction product 17, whereas R-iodo
amide 4b gave no cyclization product.
2. Effect of Bu 3Sn Cl. Next, we turned our attention
to improvement of the reaction of R-iodo amide 7c. The
fact that radical reaction of 7a using Bu3SnH gave a
better yield of cyclized product than that employing
(TMS)3SiH (Table 1, entry 1 vs 4) led us to consider the
influence of Bu3SnCl generated as the reaction of 7a with
Bu3SnH proceeds. We therefore examined radical reac-
tions of 7c in the presence of various types of Bu3SnX
(Table 2). The reaction using 1 equiv of Bu3SnCl as an
additive gave cyclization products 2, 8, and 9 in 38%
combined yield along with simple reduction product 10
in 30% yield (entry 1). We were delighted to find that
use of a large excess of Bu3SnCl improved the yield of
cyclization products to 66% (total yield) (entry 2). When
Bu3SnF or Bu3SnOAc was employed as an additive,
enamide 8 was obtained selectively (entries 3 and 4). In
contrast, the addition of Bu3SnI or (TMS)3SiCl mainly
afforded simple reduction product 10 (entries 5 and 6).20
One possible explanation of the role of Bu3SnCl as an
additive may involve consideration of initial conforma-
(20) Although there has been no sufficient work on the refinement
of Lewis acids, titanium tetraisopropoxide and diethylaluminum
chloride gave unsatisfactory results due to decomposition of the
starting material 7c. Ytterbium(III) triflate was not appropriate as
the Lewis acid because of its low solubility in toluene.
(21) An example of controlling a rotamer population with Bu3SnCl
has been reported. See: Sibi, M. P.; J i, J . J . Am. Chem. Soc. 1996,
118, 3063. For a review on the use of Lewis acids in free radical
reactions, see: Renaud, P.; Gerster, M. Angew. Chem., Int. Ed. Engl.
1998, 37, 2562. Bu3SnCl acts as a Lewis acid in regioselective reductive
cleavage of ketals. See: Srikrishna, A.; Viswajanani, R. Synlett 1995,
95.
(22) This discussion is based on the consideration that complex 7c-
Sn undergoes an attack by Bu3Sn• more rapidly than 7c. It was
reported that asymmetric radical allylation of an R-methyl-R-iododi-
hydrocoumarin with allyltributyltin in the presence of only 0.2 equiv
of a chiral Lewis acid gave the allylation product in 81% yield with
80% ee. This suggests that an R-halo carbonyl compound complexed
with
a Lewis acid exhibits higher reactivity for Bu3Sn• than an
uncomplexed one; see: Murakata, M.; J ono, T.; Mizuno, Y.; Hoshino,
O. J . Am. Chem. Soc. 1997, 119, 11713.
5540 J . Org. Chem., Vol. 67, No. 16, 2002

Effect of Halogen on 5-Endo-trig Radical Cyclization
TABLE 3. Ra d ica l Cycliza tion of r-Ha lo Am id es 19a ,b
products (% yield)
additive
entry
R-halo amide
(5 equiv)
20
21
1
2
3
4
5
19a (X ) Cl)
19a (X ) Cl)
19a (X ) Cl)
19b (X ) I)
19b (X ) I)
53
61
59
51
65
9
14
9
7
12
Bu3SnCl
Bu3SnF
Bu3SnCl
ment of 19a with Bu3SnH in the presence of AIBN in
boiling benzene gave cyclization product 20 and uncyc-
lized reduction product 21 in 53 and 9% yields, respec-
tively (entry 1). Radical reaction of 19a with Bu3SnH in
the presence of Bu3SnCl or Bu3SnF afforded better yields
of 20 (entries 2 and 3). The effect of Bu3SnCl was also
observed in the reaction of 19b. Thus, the reaction of 19b
in the absence of Bu3SnCl provided 20 and 21 in 51 and
7% yields, respectively, whereas in the presence of
Bu3SnCl, the yield of 20 was improved to 65%.23
3. F or m a l Syn th eses of (()-r-Lycor a n e a n d (()-
γ-Lycor a n e. [2-Halo-4,5-(methylenedioxy)benzyl]hexa-
(or tetra)hydroindones such as 23a and 23b are known
to be intermediates for syntheses of lycoranes 22 having
galanthane ring systems within the Amaryllidaceae
alkaloid family (Figure 3). Several groups have synthe-
sized lycoranes by employing radical cyclization or Heck
reaction of 23a or 23b.3g,4d,24 Therefore, we envisioned a
short-step entry to lycoranes using the sequential 5-endo
and 6-endo radical cyclizations from radical precursor 26,
on the basis of the aspect that radical reaction of R-iodo
amide 7c in the presence of Bu3SnF gave hexahydroin-
dolone 8, closely related to 23a , as the major product
(Table 2, entry 3).25 Bu3SnH-Mediated 5-endo radical
cyclization of R-iodo amide 26 would first provide ena-
mide 25, which in turn could undergo 6-endo radical
cyclization in situ to afford lycorane framework 24.
F IGUR E 3. Synthetic plan for lycoranes by employing
sequential 5-endo and 6-endo radical cyclizations of R-iodo
amide 26.
SCHEME 6a
The requisite R-iodo amide 26 was readily prepared
from cyclohexanone (Scheme 6). Thus, condensation of
cyclohexanone with 2-iodo-(4,5-methylenedioxy)benzyl-
amine26 followed by treatment of the resulting crude
imine 27 with chloroacetyl chloride and triethylamine
furnished R-chloro amide, which was treated with sodium
iodide to afford 26.
a
Key: (a) 2-iodo-4,5-(methylenedioxy)benzylamine, toluene,
reflux; (b) ClCOCH2Cl, Et3N, toluene, 0 °C f rt, 51% (two steps);
(c) NaI, acetone, rt, 42%.
With R-iodo amide 26 in hand, we examined the
sequential radical cyclization of 26 (Table 4). R-Iodo
amide 26 was treated with Bu3SnH (2.4 equiv) and N,N-
azobis(cyclohexanecarbonitrile) (ACN) (0.1 equiv) in the
presence of Bu3SnF (5 equiv) in boiling toluene to give
lycorane derivative 24a (21%) and its dehydro derivative
28 (25%) along with 5-endo cyclization products 29 (13%)
and 30 (18%) (entry 1). The structure of 24a was
established by direct comparison of the 1H NMR spec-
trum with that reported.24b Use of benzene for the
reaction slightly increased the total yield of double
cyclization products 24a , 24b, and 28 (entry 2). Use of
xylene as a solvent or Bu3SnOAc as an additive did not
result in any improvement in the reaction (entries 3 and
4). Since the products 24a and 24b had already been
(23) The possibility of coordination of Bu3SnCl with the ester group
of 19 cannot be ruled out, whereas the amide carbonyl group seems to
be a stronger Lewis base than the ester carbonyl group. Therefore, it
is reasonable to consider that the mechanism of the reaction of 19 in
the presence of Bu3SnX is similar to that of 7. The reason for the
increased yield of the reduction product 21 in the presence of Bu3-
SnCl is, however, still obscure at the moment.
(24) (a) Yoshizaki, H.; Satoh, H.; Sato, Y.; Nukui, S.; Shibasaki, M.;
Mori, M. J . Org. Chem. 1995, 60, 2016. (b) Rigby, J . H.; Mateo, M. E.
Tetrahedron 1996, 52, 10569. See also: (c) Cossy, J .; Tresnard, L.; Pard,
D. G. Eur. J . Org. Chem. 1999, 1925. (d) Padwa, A.; Brodney, M. A.;
Lynch, S. M. J . Org. Chem. 2001, 66, 1716.
(25) For
a synthesis of (()-γ-lycorane using a cascade reaction
initiated by an acylaminyl radical, see: Hoang-Cong, X.; Quiclet-Sire,
B.; Zard, S. Z. Tetrahedron Lett. 1999, 40, 2125.
(26) Beugelmans, R.; Chastanet, J .; Ginsburg, H.; Quintero-Cortes,
L.; Roussi, G. J . Org. Chem. 1985, 50, 4933.
J . Org. Chem, Vol. 67, No. 16, 2002 5541

Tamura et al.
TABLE 4. Ra d ica l Cycliza tion of r-Iod o Am id e 26
products (% yield)
conditionsa
entry
additive (5 equiv)/solvent
24a
24b
28
29
30
1
2
3
4
Bu3SnF/toluene
Bu3SnF/benzene
Bu3SnF/xylene
Bu3SnOAc/xylene
21
29
16
18
25
27
20
21
13
18
10
18
15
5
4
6
15
17
a
All reactions were carried out by using Bu3SnH (2.4 equiv) and ACN (0.1 equiv) under refluxing conditions.
SCHEME 7
previous result with the radical from 23b (X ) Br), which
undergoes cyclization in a 6-exo manner to afford a
γ-lycorane-type product.3g
Con clu sion
We have revealed that the general guideline that “an
iodine atom acts as a better leaving group for Bu3SnH-
mediated radical cyclization of ω-halo alkenes” is not
applicable to the 5-endo-trig cyclization of R-halo amides
having an alkenic sp2 carbon atom R to the amide
nitrogen atom. In addition, we also showed that the
cyclization ability of R-iodo amide can be restored by the
use of Bu3SnCl or Bu3SnF as an additive. Moreover, this
cyclization could be applied to a short-step synthesis of
(()-lycoranes.
Exp er im en ta l Section
Melting points are uncorrected. Column chromatography
was performed on silica gel 60 PF254 under pressure.
N-Ben zyl-2-ch lor o-N-(cycloh ex-2-en yl)a ceta m id e (1a ).
To a stirred solution of 3-bromocyclohex-1-ene (1.79 g, 10
mmol) and benzylamine (3.21 g, 30 mmol) in CH3CN (30 mL)
was added K2CO3 (1.3 g, 20 mmol) at room temperature, and
the mixture was further stirred for 1 h. The mixture was
diluted with water and extracted with AcOEt. The organic
phase was dried (Na2SO4) and concentrated under reduced
pressure. The residue was chromatographed (1:6 n-hexane/
AcOEt) to give N-benzyl-N-(cyclohex-2-enyl)amine (1.13 g,
60%). This material was used in the next step without further
purification. To a solution of the amine (935 mg, 5 mmol) and
triethylamine (0.83 mL, 6 mmol) in benzene (10 mL) was added
a solution of chloroacetyl chloride (678 mg, 6 mmol) in benzene
(2 mL) at room temperature, and the mixture was stirred for
2 h. The mixture was washed with water, dried (MgSO4), and
concentrated under reduced pressure. The residue was chro-
matographed (3:1 n-hexane/AcOEt) to afford 1a (1.30 g, 99%)
as an oil. Two rotamers of 1a (3:2) due to the amide bond were
converted to R-lycorane (22a)24b and γ-lycorane (22b),3g,4d,24a
respectively, a short-step synthesis of 22a and 22b could
be realized by the present reaction, although the yields
were moderate.
Formation of 24a ,b and 28 may involve 6-endo cycliza-
tion of aryl radical 31 generated from intermediate 25
(Scheme 7). An attack of aryl radical 31 on the enamide
double bond from the less-hindered â-face gives R-ami-
doyl radical 32a , and R-face attack gives 32b. The
radicals 32a and 32b undergo attacks of Bu3SnH to
afford 24a and 24b, respectively, and SET or atom
transfer reaction from R-amidoyl radicals 32a ,b to start-
ing R-iodo amide 26 in a way similar to that of 8 affords
28. Simple reduction of the radical 31 furnishes 29. It is
interesting to note that 6-endo cyclization of 31 gives
R-lycorane derivative 24a 27 predominantly, in light of our
1
observed by H and 13C NMR: 1H NMR (270 MHz, CDCl3) δ
1.36-1.99 (m, 6 H), 3.86 (s, 2 H × 3/5), 4.20 (s, 2 H × 2/5),
4.34 (d of a pair of ABq, J ) 15.6 Hz, 1 H × 2/5), 4.47-4.55
(m, 1 H × 2/5), 4.51 (d of a pair of ABq, J ) 18.2 Hz, 1 H ×
3/5), 4.60 (d of a pair of ABq, J ) 18.2 Hz, 1 H × 3/5), 4.69 (d
of a pair of ABq, J ) 15.6 Hz, 1 H × 2/5), 5.28-5.33 (m, 1 H
× 3/5), 5.47 (t, J ) 9.4 Hz, 1 H), 5.91 (dt, J ) 9.4, 2.9 Hz, 1 H),
7.18-7.38 (m, 5 H); 13C NMR (67.8 MHz, CDCl3) δ 21.2 (major),
21.5 (minor), 24.3 (minor), 24.6 (major), 27.4 (major), 29.0
(minor), 41.6 (minor), 42.1 (major), 46.4 (minor), 47.2 (major),
(27) Rigby and co-workers also reported that a 6-endo aryl radical
cyclization similar to that of radical 31 affords an R-lycorane-type
product. See ref 24b.
5542 J . Org. Chem., Vol. 67, No. 16, 2002

Effect of Halogen on 5-Endo-trig Radical Cyclization
52.3 (major), 56.1 (minor), 125.5 (major), 126.7 (minor), 126.9
(major), 127.0 (minor), 127.50 (major), 128.3 (minor), 129.0
(major), 132.5 (minor), 132.8 (major), 138.1 (major), 138.5
128.1, 128.2, 128.7, 131.6, 132.0, 138.9, 139.5, 170.9, 171.9;
HRMS calcd for C15H19NO 229.1466, found 229.1470. (b)
Ra d ica l Rea ction of 1b. According to the general procedure,
a boiling solution of 1b (177 mg) in toluene was treated with
a solution of Bu3SnH and AIBN in toluene (4 h for addition).
Workup and chromatography gave 2 (60 mg, 53%) and 3 (7
mg, 6%).
(minor), 167.2 (minor), 167.6 (major). Anal. Calcd for C15H18
-
ClNO: C, 68.30; H, 6.88; N, 5.31. Found: C, 68.18; H, 6.95; N
5.35.
N-Ben zyl-N-(cycloh ex-2-en yl)-2-iod oa ceta m id e (1b). To
a stirred solution of 1a (480 mg, 1.8 mmol) in acetone (10 mL)
was added NaI (540 mg, 3.6 mmol) at room temperature, and
the mixture was further stirred for 15 h. The mixture was
diluted with water and extracted with Et2O. The organic phase
was washed successively with a 10% solution of Na2S2O3 and
brine, dried (MgSO4), and concentrated under reduced pres-
sure. The residue was chromatographed (1:5 Et2O/benzene)
to afford 1b (224 mg, 47%) as a colorless oil. Two rotamers of
1b (3:2) due to the amide bond were observed by 1H NMR: 1H
NMR (270 MHz, CDCl3) δ 1.25-1.99 (m, 6 H), 3.52 (s, 2 H ×
3/5), 3.85 (d of a pair of ABq, J ) 9.9 Hz, 1 H × 2/5), 3.90 (d
of a pair of ABq, J ) 9.9 Hz, 1H × 2/5), 4.33 (d of a pair of
ABq, J ) 15.8 Hz, 1 H × 2/5), 4.41-4.48 (m, 1 H × 2/5), 4.47
(d of a pair of ABq, J ) 18.1 Hz, 1 H × 3/5), 4.57 (d of a pair
of ABq, J ) 18.1 Hz, 1 H × 3/5), 4.67 (d of a pair of ABq, J )
15.8 Hz, 1 H × 2/5), 5.26-5.32 (m, 1 H × 3/5), 5.45 (dq, J )
10.2, 1.6 Hz, 1 H × 3/5), 5.52 (dq, J ) 10.2, 1.6 Hz, 1 H × 2/5),
5.91 (dq, J ) 9.9, 3.3 Hz, 1 H), 7.18-7.39 (m, 5H). These
spectral data were identical with those reported.28
Gen er a l P r oced u r e for Ra d ica l Rea ction s. To a boiling
solution of R-halo amide (0.5 mmol) in toluene (or benzene)
(50 mL) was added dropwise a solution of Bu3SnH (175 mg,
0.60 mmol) [or (TMS)3SiH (149 mg, 0.60 mmol)] and AIBN (8
mg, 50 µmol) [or ACN (10 mg, 50 µmol)] in toluene (or ben-
zene) (50 mL) over 1-4 h by employing a syringe-pump
technique, and the mixture was further heated at reflux for
an appropriate period of time. After evaporation of the solvent,
Et2O (20 mL) and an 8% aqueous KF solution (50 mL) were
added to the residue, and the mixture was vigorously stirred
at room temperature for 16 h. The organic phase was sepa-
rated, and the aqueous phase was further extracted with Et2O.
The combined organic phase was washed with brine, dried
(MgSO4), and concentrated. The residue was chromatographed
on silica gel (n-hexane/AcOEt) to afford the product(s).
cis-1-Ben zyl-octa h yd r oin d ol-2-on e (2) a n d N-Ben zyl-
N-(cycloh ex-2-en yl)a ceta m id e (3): (a ) Ra d ica l Rea ction
of 1a . According to the general procedure, a boiling solution
of chloro amide 1a (132 mg) in toluene was treated with a
solution of Bu3SnH and AIBN in toluene (4 h for addition; 3 h
further for heating). Since 1a was not consumed, an addi-
tional solution of Bu3SnH (175 mg, 0.60 mmol) and AIBN (8
mg, 0.05 mmol) in toluene was added to the mixture over 3 h,
and the mixture was further heated for 7 h. After workup, the
crude material was chromatographed on silica gel (3:1 n-
hexane/AcOEt) to afford 228 (48 mg, 41%) and 329 (17 mg,
15%). 2: 1H NMR (270 MHz) δ 1.22-1.51 (5 H, m), 1.54-1.70
(3 H, m), 2.17-2.44 (3 H, m), 3.39 (1 H, q, J ) 5.6 Hz), 3.97 (1
H, d of a pair of ABq, J ) 15.0 Hz), 4.93 (1 H, d of a pair of
ABq, J ) 15.0 Hz), 7.21-7.33 (5 H, m); 13C NMR (67.8
MHz, CDCl3) δ 21.3, 22.6, 24.7, 27.2, 27.9, 49.9, 127.4, 128.3,
128.7, 129.2, 137.2, 137.8, 166.1. These spectral data were
identical with those reported.28 3: two rotamers of 3 (1:1)
were observed by the 1H and 13C NMR spectra due to the
amide bond; 1H NMR (270 MHz, CDCl3) δ 1.26-2.06 (m, 6 H),
2.77 (s, 3 H × 1/2), 2.79 (s, 3 H × 1/2), 3.72 (s, 2 H × 1/2)
3.76 (s, 2 H × 1/2), 4.44 (m, 1 H × 1/2), 5.21-5.46 (m, 1 H +
1 H × 1/2), 5.84-5.92 (m, 1 H), 7.21-7.33 (m, 5 H); 13C NMR
(67.8 MHz, CDCl3) δ 21.3, 21.6, 22.1, 22.4, 24.4, 24.6, 27.7,
28.8, 46.0, 48.1, 51.3, 56.1, 125.6, 126.5, 127.0, 127.1, 128.0,
N-Cycloh ex-1-en yl-N-m eth yl-2-iod oa ceta m id e (4b). Ac-
cording to a procedure similar to that for the preparation of
1b, chloro amide 4a 3b (225 mg, 1.2 mmol) was treated with
NaI (270 mg, 1.8 mmol) in dry acetone (30 mL). After workup,
the crude material was chromatographed (2:1 n-hexane/
AcOEt) to give 4b (335 mg, quantitative) as a colorless oil: 1H
NMR (270 MHz, CDCl3) δ 1.58-1.67 (m, 2 H), 1.74-1.83 (m,
2 H), 2.12-2.19 (m, 2 H), 2.21-2.24 (m, 2 H), 2.98 (s, 3 H),
3.81 (s, 2 H), 5.85 (br s, 1 H); 13CNMR (67.8 MHz, CDCl3) δ
-2.6, 24.0, 25.2, 27.2, 37.3, 129.2, 142.8, 169.7; HRMS calcd
for C9H14NOI 279.0220, found 279.0115. Anal. Calcd for
C9H14NOI: C, 38.73; H, 5.06; N, 5.02. Found: C, 38.59; H, 5.01;
N, 5.06.
N-(Cycloh ex-1-en yl)-N-m eth yla ceta m id e (6). According
to the general procedure, a boiling solution of 4b (140 mg) in
toluene was treated with a solution of Bu3SnH and AIBN in
toluene for 2 h. Workup and chromatography (8:1 n-hexane/
AcOEt) gave 63b (42 mg, 54%) as a colorless oil: 1H NMR (270
MHz, CDCl3) δ 1.55-1.64 (m, 2 H), 1.69-1.78 (m, 2 H), 2.01
(s, 3 H), 2.04-2.15 (m, 4 H), 2.97 (s, 3 H), 5.64 (br s, 1 H).
This 1H NMR spectrum was identical with that reported for
6.3b
N-Ben zyl-2-ch lor o-N-(cycloh ex-1-en yl)a ceta m id e (7a ).
A mixture of cyclohexanone (981 mg, 10 mmol) and benzyl-
amine (1.07 g, 10 mmol) in toluene (50 mL) was heated at
reflux with a Dean-Stark trap for 2 h. After cooling, the
mixture was concentrated to ca. 25 mL under reduced pres-
sure. The mixture was added dropwise to a stirred solution of
chloroacetyl chloride (2.26 g, 20 mmol) in toluene (50 mL) at
0 °C, and the mixture was further stirred under the same
temperature for 2 h. To the mixture was added triethylamine
(4.1 mL, 30 mmol) at 0 °C, and the mixture was further stirred
under the same temperature for 2 h. The mixture was washed
successively with a saturated aqueous solution of NaHCO3 and
with brine. After drying (MgSO4), the mixture was concen-
trated and purified by column chromatography on silica gel
(5:1 n-hexane/AcOEt) to give 7a (1.22 g, 46%) as a colorless
oil: 1H NMR (270 MHz, CDCl3) δ 1.52-1.70 (m, 4 H), 2.03 (br
s, 4 H), 4.13 (s, 2 H), 4.63 (s, 2 H), 5.47 (br s, 1 H), 7.26-7.30
(m, 5 H); 1H NMR (270 MHz, C6D6) δ 1.05-1.20 (m, 2 H),
1.20-1.32 (br s, 2 H), 1.52 (br s, 1 H), 1.60 (br s, 1 H), 3.85 (s,
2 H), 4.49 (s, 2 H), 4.99 (br s, 1 H), 7.00-7.28 (m, 5 H); 13C
NMR (67.8 MHz, CDCl3) δ 21.3, 22.6, 24.7, 27.9, 41.7, 49.9,
127.4, 128.3, 128.7, 129.3, 137.1, 137.4, 165.8. Anal. Calcd for
C
15H18ClNO: C, 68.30; H, 6.88; N, 5.31. Found: C, 68.18; H,
6.92; N, 5.29.
N-Ben zyl-2-br om o-N-(cycloh ex-1-en yl)a ceta m id e (7b).
According to a procedure similar to that described for the
preparation of 7a , treatment of cyclohexanone (981 mg, 10
mmol) with benzylamine (1.07 g, 10 mmol) followed by
treatment of the resulting imine with bromoacetyl bromide
(2.20 g, 11 mmol) and triethylamine (4.1 mL, 30 mmol) gave
crude 7b, which was chromatographed (5:1 n-hexane/AcOEt)
to afford 7b (1.22 g, 46%) as a colorless oil: 1H NMR (270 MHz,
CDCl3) δ 1.52-1.70 (m, 4 H), 2.03-2.07 (m, 4 H), 3.94 (s, 2
H), 4.62 (s, 2 H), 5.52 (br s, 1 H), 7.27-7.30 (m, 5 H); 1H NMR
(270 MHz, C6D6) δ 1.05-1.20 (m, 2 H), 1.20-1.38 (m, 2 H),
1.55 (br s, 2 H), 1.68 (br s, 2 H), 3.67 (s, 2 H), 4.49 (2 H, s),
5.10 (br s, 1 H), 7.00-7.30 (m, 5 H); 13C NMR (67.8 MHz,
CDCl3) δ 21.3, 22.6, 24.7, 27.2, 27.9, 49.9, 127.4, 128.3, 128.7,
129.1, 137.2, 137.8, 166.1. Anal. Calcd for C15H18BrNO: C,
58.45; H, 5.89; N, 4.54. Found: C, 58.17; H, 5.93; N 4.19.
(28) Mori, M.; Kanda, N.; Oda, I.; Ban, Y. Tetrahedron 1985, 41,
5465.
(29) Mori, M.; Oda, I.; Ban, Y. Tetrahedron Lett. 1982, 23, 5315.
(30) Erhanov, K. B.; Kayunova, L. A.; Yakimovich, S. I.; Umarova,
Z. N.; Krasnomolova, L. P. Zh. Org. Khim. 1983, 19, 2329; Chem. Abstr.
1984, 100, 138355.
N-Ben zyl-N-(cycloh ex-1-en yl)-2-iodoacetam ide (7c). Ac-
cording a procedure similar to that described for the prepara-
tion of 1b, chloro amide 7a (526 mg, 2 mmol) was treated with
J . Org. Chem, Vol. 67, No. 16, 2002 5543

Tamura et al.
NaI (330 mg, 2.2 mmol) in dry acetone (30 mL). After workup,
the crude material was chromatographed (5:1 n-hexane/AcOEt)
to give 7c (732 mg, quantitative) as a colorless oil: 1H NMR
(270 MHz, CDCl3) δ 1.49-1.58 (m, 2 H), 1.64-1.76 (m, 2 H),
2.00-2.04 (m, 2 H), 2.11 (br s, 2 H), 3.85 (s, 2 H), 4.60 (s, 2 H),
5.57 (br s, 1 H), 7.21-7.32 (m, 5 H); 1H NMR (270 MHz, C6D6)
δ 1.05-1.20 (m, 2 H), 1.20-1.35 (m, 2 H), 1.52-1.67 (m, 2 H),
1.75 (br s, 2 H), 3.56 (s, 2 H), 4.50 (s, 2 H), 5.20 (br s, 1 H),
7.00-7.31 (m, 5 H); 13C NMR (67.8 MHz, CDCl3) δ -2.3, 21.3,
22.7, 24.6, 27.6, 50.0, 127.3, 128.3, 128.6, 128.9, 137.4, 138.2,
167.3. Anal. Calcd for C15H18INO: C, 50.72; H, 5.11; N, 3.94.
Found: C, 50.65; H, 5.13; N 3.62.
imine was dissolved in toluene (20 mL). To this solution were
added successively pyridine (2.4 mL, 30 mmol) and a solution
of 2-bromopropanoyl bromide (4.75 g, 22 mmol) in toluene (50
mL) at 0 °C, and the mixture was stirred under the same
temperature for 1 h. The mixture was washed successively
with 5% hydrochloric acid, a saturated aqueous solution of
NaHCO3, and brine. After drying (MgSO4), the mixture was
concentrated under reduced pressure. The residue was purified
by column chromatography on silica gel (5:1 n-hexane/AcOEt)
to yield 2-bromo-N-(cyclohex-1-enyl)-N-methylpropionamide.
This material was used in the next step without further
purification. According to a procedure similar to that described
for the preparation of 1b, the bromo amide (550 mg, 1.9 mmol)
was treated with NaI (330 mg, 2.2 mmol) in dry acetone (30
mL). After workup, the crude material was chromatographed
(3:1 n-hexane/AcOEt) to give 15b (460 mg, 8%, two steps) as
a colorless oil: 1H NMR (270 MHz, CDCl3) δ 1.58-1.82 (m, 4
H), 1.93 (d, J ) 6.9 Hz, 3 H), 2.04-2.31 (m, 4 H), 2.98 (s, 3 H),
4.83 (q, J ) 6.9 Hz, 1 H), 5.81 (br s, 1 H); 13C NMR (67.8 MHz,
CDCl3) δ 14.4, 21.3, 22.6, 24.3, 24.5, 26.6, 34.7, 126.5, 139.8,
170.3. Anal. Calcd for C10H16INO: C, 40.97; H, 5.50; N, 4.78.
Found: C, 40.76; H, 5.65; N, 4.62.
Ben zyl-3-m eth yl-1,4,5,6,7,7a-h exah ydr oin dol-2-on e (16b)
a n d N-(Cycloh ex-1-en yl)-N-m et h ylp r op ion a m id e (17).
According to the general procedure, a boiling solution of 15b
(182 mg, 0.62 mmol) in toluene (20 mL) was treated with
Bu3SnH (198 mg, 0.68 mmol) and AIBN (10 mg, 0.062 mmol)
in toluene (45 mL) for 2 h. Workup and chromatography (from
20:1 to 1:5 n-hexane/AcOEt) gave 16b (20 mg, 24%) and 17
(53 mg, 63%). 16b: 1H NMR (270 MHz, CDCl3) δ 0.95 (qd, J
) 11.9, 3.3 Hz, 1 H), 1.26 (tt, J ) 13.2, 3.6 Hz, 1 H), 1.48 (qt,
J ) 13.2, 3.0 Hz, 1 H), 1.78 (t, J ) 1.5 Hz, 3 H), 1.82-2.12 (m,
3 H), 2.37-2.42 (m, 1 H), 2.69-2.79 (m, 1 H), 2.95 (s, 3 H),
3.50 (dd, J ) 11.2, 5.3 Hz, 1 H); HRMS calcd for C10H15NO
165.1154, found 165.1160. 17: 1H NMR (270 MHz, CDCl3) δ
1.11 (t, J ) 7.4 Hz, 3 H), 1.58-1.78 (m, 4 H), 2.08-2.14 (m, 4
H), 2.28 (q, J ) 7.4 Hz, 2 H), 2.98 (s, 3 H), 5.63 (br s, 1 H);
HRMS calcd for C10H17NO 167.1311, found 167.1313.
Meth yl N-Ben zylp yr oglu ta m a te (20) a n d Meth yl 2-(N-
Ben zyleth a n a m id o)p r op en oa te (21) (Ta ble 3, En tr y 5).
According to the general procedure, a boiling solution of 19b
(180 mg) and Bu3SnCl (813 mg, 2.5 mmol) in benzene was
treated with Bu3SnH (160 mg, 0.55 mmol) and AIBN (12 mg,
0.1 mmol) in benzene for 5 h. After workup, the crude material
was chromatographed (2:1 n-hexane/AcOEt) to give 203k (69
mg, 59%) and 213k (11 mg, 9%). 20: 1H NMR (270 MHz, CDCl3)
δ 2.02-2.63 (m, 4 H), 3.66 (s, 3 H), 4.01 (d, J ) 15.3 Hz, 1 H),
4.05 (dd, J ) 12.2, 8.9 Hz, 1 H), 5.00 (d, J ) 15.3 Hz, 1 H),
7.19-7.31 (m, 5 H). 21: 1H NMR (270 MHz, CDCl3) δ 2.00 (s,
3 H), 3.78 (s, 3 H), 4.68 (s, 2 H), 5.40 (s, 1 H), 6.32 (s, 1 H),
7.26 (m, 5 H). The 1H NMR spectral data of 20 and 21 were
identical with those reported.3k
N-(Cycloh ex-1-en yl)-N-[(2-iod o-4,5-m et h ylen ed ioxy)-
ben zyl]-2-iod oa ceta m id e (26). According to a procedure
similar to that described for the preparation of 7a , N-cyclohex-
1-enyl-N-[(2-iodo-4,5-methylenedioxy)benzyl]-2-chloroaceta-
mide (1.12 g, 52%) was obtained from cyclohexanone (490 mg,
5 mmol), 2-iodo-4,5-methylendioxybenzylamine26 (1.38 g, 5
mmol), triethylamine (4.1 mL, 30 mmol), and toluene (50 mL)
after chromatography on silica gel (10:1 n-hexane/AcOEt). 1H
NMR (270 MHz, CDCl3) δ 1.56-1.62 (m, 2 H), 1.70-1.75 (m,
2 H), 2.03-2.10 (m, 4 H), 4.13 (s, 2 H), 4.70 (s, 2 H), 5.48 (br
s, 1 H), 5.96 (s, 2 H), 6.98 (s, 1 H), 7.18 (s, 1 H). Anal. Calcd
for C16H17ClINO3: C, 44.31; H, 3.95; N, 3.23. Found: C, 44.16;
H, 3.97; N, 3.18. According to a treatment similar to that
described for the preparation of 7c, 26 (552 mg, 42%) was
obtained from the corresponding chloro amide (1.12 g, 2.5
mmol), NaI (2.0 g), and acetone (20 mL) after chromatography
on silica gel (10:1 n-hexane/AcOEt): 1H NMR (270 MHz,
CDCl3) δ 1.55-1.63 (m, 2 H), 1.70-1.79 (m, 2 H), 2.04-2.09
(s, 2 H), 2.19 (br s, 2 H), 3.84 (s, 2 H), 4.67 (s, 2 H), 5.60 (br s,
1 H), 5.96 (s, 2 H), 6.99 (s, 1 H), 7.18 (s, 1 H). Anal. Calcd for
Com p ou n d 2, 1-Ben zyl-1,3,3a ,4,5,6-h exa h yd r oin d ol-2-
on e (8), 1-Ben zyl-1,4,5,6,7,7a -h exa h yd r oin d ol-2-on e (9),
an d N-(Cycloh ex-1-en yl)-N-m eth ylacetam ide (10) (Tables
1 a n d 2): (a ) Ta ble 1, En tr y 1. According to the general
procedure, a solution of 7a (132 mg) in toluene was treated
with a solution of Bu3SnH and AIBN in toluene for 4.5 h.
Workup and chromatography (10:1 n-hexane/AcOEt) gave 2
(106 mg, 92%) as a colorless oil. The 1H NMR spectral data
were identical with those obtained from radical reaction of 1a .
(b) Ta ble 1, En tr y 2. According to a procedure similar to that
described above in a , R-bromo amide 7b (154 mg) was treated
with Bu3SnH to give 2 (63 mg, 55%), 8 (12 mg, 11%), and 94b
(13 mg, 11%). 8: 1H NMR (270 MHz, CDCl3) δ 1.25-1.65 (m,
2 H), 1.82-2.14 (m, 4 H), 2.20 (dd, J ) 15.4, 9.5 Hz, 1 H),
2.62-2.78 (m, 2 H), 4.47 (d of a pair of ABq, J ) 15.5 Hz, 1
H), 4.76 (d of a pair of ABq, J ) 15.5 Hz, 1 H), 4.77 (dd, J )
5.6, 3.3, 1 H), 7.21-7.36 (m, 5 H); HRMS calcd for C15H17NO
227.1312, found 227.1308. 9: 1H NMR (270 MHz, CDCl3) δ
0.92-1.11 (m, 1 H), 1.19-1.43 (m, 2 H), 1.17-2.39 (m, 4 H),
2.72 (br d, J ) 11.2 Hz, 1 H), 3.58 (dd, J ) 11.2, 5.9 Hz, 1 H),
4.17 (d, J ) 15.1 Hz, 1 H), 4.99 (d, J ) 15.1 Hz, 1 H), 5.80 (s,
1 H), 7.21-7.37 (m, 5 H). The 1H NMR spectral data were
identical with those reported. 4b (c) Ta ble 1, En tr y 3. Accord-
ing to a procedure similar to that described above in a , R-iodo
amide 7c (178 mg, 0.50 mmol) was treated with Bu3SnH to
give 8 (15 mg, 13%), 9 (12 mg, 11%), and 1030 (78 mg, 68%).
10: 1H NMR (270 MHz, CDCl3) δ 1.47-1.69 (m, 4 H), 1.97-
2.03 (m, 4 H), 2.06 (s, 3 H), 4.61 (s, 2 H), 5.39 (br s, 1 H), 7.22-
7.28 (m, 5 H); 13C NMR (67.8 MHz, CDCl3) δ 21.4, 21.5, 22.7,
24.6, 28.0, 49.4, 127.0, 128.0, 128.1, 128.6, 138.0, 138.8, 169.8;
HRMS calcd for C17H19NO 229.1467, found 229.1467. (d )
Ta ble 1, En tr y 4. According to the general procedure, a
boiling solution of 7a (132 mg) in toluene was treated with a
solution of (TMS)3SiH and AIBN in toluene for 8 h. After
evaporation, the residue was chromatographed (4:1 n-hexane/
AcOEt) to afford 2 (65 mg, 56%) and 9 (14 mg, 12%). (e) Ta ble
1, En tr y 5. According to a procedure similar to that described
above in d , a boiling solution of 7c (178 mg, 0.50 mmol) in
toluene was treated with a solution of (TMS)3SiH and AIBN
in toluene for 4 h. To the mixture was added a saturated
solution of NaHCO3, and the mixture was stirred vigorously
for 1 h. The mixture was extracted with AcOEt, dried (MgSO4),
and concentrated under reduced pressure. The residue was
chromatographed (3:1 n-hexane/AcOEt) to give 10 (87 mg,
76%). (f) Ta ble 2, En tr y 2. According to the general proce-
dure, a boiling solution of 7c (178 mg) and Bu3SnCl (813 mg,
2.5 mmol) in toluene was treated with a solution of Bu3SnH
and AIBN in toluene for 1 h. Workup and chromatography
gave 2 (21 mg, 18%), 8 (23 mg, 20%), 9 (32 mg, 28%), and 10
(10 mg, 9%). (g) Ta ble 2, En tr y 3. According to the general
procedure, a boiling solution of 7c (178 mg) and Bu3SnF (771
mg, 2.5 mmol) in toluene was treated with a mixture of
Bu3SnH and AIBN (12 mg, 0.1 mmol) in toluene for 2 h.
Workup and chromatography gave 2 (10 mg, 9%) and 8 (61
mg, 54%).
N -(Cycloh e x-1-e n yl)-N -m e t h yl-2-iod op r op ion a m id e
(15b). A mixture of cyclohexanone (2.1 mL, 20 mmol) and
methylamine (3 mL) in toluene (10 mL) was heated in a sealed
tube at 100 °C for 2 h. After cooling, the mixture was
concentrated under reduced pressure, and the residual crude
5544 J . Org. Chem., Vol. 67, No. 16, 2002

Effect of Halogen on 5-Endo-trig Radical Cyclization
16H17I2NO3: C, 36.67; H, 3.30; N, 2.67. Found: C, 36.67; H,
C
98.4, 101.4, 108.4, 108.6, 121.2, 131.0, 142.3, 147.2, 148.3,
175.1; HRMS calcd for C16H17NO3 271.1209, found 271.1209.
30: 1H NMR (270 MHz, CDCl3) δ 1.25-1.75 (m, 7 H), 2.01-
2.40 (m, 4 H), 3.39 (q, J ) 5.6 Hz, 1 H), 3.87 (d, J ) 14.8 Hz,
1 H), 4.84 (d, J ) 14.8 Hz, 1 H), 5.94 (s, 2 H), 6.70-6.75 (m,
3 H); 13C NMR (67.8 MHz, CDCl3) δ 21.1, 22.3, 26.7, 27.2, 32.3,
37.0, 43.6, 55.9, 101.0, 108.1, 108.4, 121.2, 130.9, 146.8, 147.9,
175.5; HRMS calcd for C16H19NO3 273.1365, found 273.1366.
(b) Ta ble 4, En tr y 2. According to the general procedure, a
solution of 26 (263 mg) and Bu3SnF (771 mg, 2.5 mmol) in
benzene was treated with Bu3SnH and ACN in benzene for
10 h. Workup and chromatography gave 24a (40 mg, 29%),
24b3g (7 mg, 5%), 28 (37 mg, 27%), and 30 (14 mg, 10%). 24b:
1H NMR (270 MHz, CDCl3) δ 1.10-1.38 (m, 3 H), 1.67-1.76
(m, 3 H), 2.10 (d, J ) 16.1 Hz, 1 H), 2.41 (quintet, J ) 5.6 Hz,
1 H), 2.58 (dd, J ) 16.1, 6.8 Hz, 1 H), 2.75 (dt, J ) 12.2, 4.6
Hz, 1 H), 3.77 (t, J ) 4.6 Hz, 1 H), 4.32 (d of a pair of ABq, J
) 17.1 Hz, 1 H), 4.54 (d of a pair of ABq, J ) 17.1 Hz, 1 H),
5.92 (d of a pair of ABq, J ) 1.4 Hz, 1 H), 5.94 (d of a pair of
ABq, J ) 1.4 Hz, 1 H), 6.59 (s, 1 H), 6.62 (s, 1 H). The 1H
NMR spectral data were identical with those reported for
24b.3g
3.30; N 2.58.
(3a R*,11bR*,11cS*)-9,10-(Meth ylen ed ioxy)-1,2,3,3a ,4,5,-
11b ,11c-oct a h yd r op yr r olo[3,2,1-d e]p h en a n t h r id in e-5-
on e (24a), Its (11bS*)-Isom er (24b), 9,10-(Meth ylen edioxy)-
1,2,3,3a ,4,5-h exa h yd r op yr r olo[3,2,1-d e]p h en a n th r id in e-
5-on e (28), 1-(3,4-Meth ylen edioxy)ben zyl-1,3,3a,4,5,6-h exa-
h yd r oin d ole-2-on e (29), a n d 1-[(3,4-Meth ylen ed ioxy)ben -
zyl]octa h yd r oin d ol-2-on e (30): (a ) Ta ble 4, En tr y 1.
According to the general procedure, a solution of 26 (263 mg)
and Bu3SnF (772 mg, 2.5 mmol) in toluene was treated with
a solution of Bu3SnH and ACN in toluene for 10 h. After
workup, the crude material was chromatographed (10:1 n-
hexane/AcOEt) to give 24a 24b (28 mg, 21%), 28 (33 mg, 25%),
29 (18 mg, 13%), and 30 (25 mg, 18%). 24a : 1H NMR (270
MHz, CDCl3) δ 1.16-1.25 (m, 1 H), 1.58-1.65 (m, 1 H), 1.72-
1.80 (m, 3 H), 2.13-2.30 (m, 2 H), 2.35-2.50 (m, 1 H), 2.50
(dd, J ) 16.5, 11.2 Hz, 1 H), 2.56-2.75 (m, 1 H), 3.21 (dd, J )
10.5, 7.2 Hz, 1 H), 4.19 (d, J ) 17.1 Hz, 1 H), 4.97 (d, J ) 17.1
Hz, 1 H), 5.91-5.93 (m, 2 H), 6.59 (s, 1 H), 6.69 (s, 1 H). The
1H NMR spectral data were identical with those reported for
24a .24b 28: 1H NMR (270 MHz, CDCl3) δ 1.45-1.89 (m, 6 H),
2.25 (dd, J ) 15.5, 7.6 Hz, 1 H), 2.49-2.59 (m, 1 H), 2.83 (t, J
) 14.1 Hz, 1 H), 4.18 (d, J ) 14.9 Hz, 1 H), 4.72 (d, J ) 14.9
Hz, 1 H), 5.97 (s, 2 H), 6.70 (s, 1 H), 6.86 (s, 1 H); HRMS calcd
for C16H15NO3 269.1052, found 269.1044. 29: 1H NMR (270
MHz, CDCl3) δ 1.25-1.66 (m, 3 H), 1.83-2.11 (m, 4 H), 2.14
(dd, J ) 15.5, 5.9 Hz, 1 H), 2.55-2.67 (m, 2 H), 4.38 (d of a
pair of ABq, J ) 15.1 Hz, 1 H), 4.66 (d of a pair of ABq, J )
15.1 Hz, 1 H), 4.79 (m, 1 H), 5.93 (s, 2 H), 6.72 (s, 3 H); 13C
NMR (67.8 MHz, CDCl3) δ 22.7, 23.5, 28.4, 35.2, 37.2, 43.6,
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra for
8, 16b, 17, and 28; 13C NMR spectra for 3, 4b, 10, 29, and 30;
and IR spectra for 1a , 4b, 7a -c, 8, 10, 15b, 26, and 29. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O020007U
J . Org. Chem, Vol. 67, No. 16, 2002 5545