Six- versus five-membered ring formation in radical cyclization of 1-vinyl-5-methyl-5-hexenyl radicals

Article abstract of DOI:10.1016/S0040-4039(02)00866-3

Radical cyclization of 1-vinyl-5-methyl-5-hexenyl radicals (radical numbering) affords six-membered ring products prevailing over the isomeric five-membered ring compounds; the former are generated through two reaction pathways: 6-endo-trig ring closure a

Full text of DOI:10.1016/S0040-4039(02)00866-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 4997–5000
Six- versus five-membered ring formation in radical cyclization
of 1-vinyl-5-methyl-5-hexenyl radicals
Ana M. Go´mez,* Mar´ıa D. Company, Clara Uriel, Seraf´ın Valverde and J. Cristo´bal Lo´pez*
Instituto de Qu´ımica Orga´nica General, C.S.I.C., Juan de la Cierva 3, 28006 Madrid, Spain
Received 4 April 2002; revised 26 April 2002; accepted 7 May 2002
Abstract—Radical cyclization of 1-vinyl-5-methyl-5-hexenyl radicals (radical numbering) affords six-membered ring products
prevailing over the isomeric five-membered ring compounds; the former are generated through two reaction pathways: 6-endo–trig
ring closure and rearrangement of intermediate methylenecyclopentyl radicals obtained by 5-exo–trig cyclization. © 2002 Elsevier
Science Ltd. All rights reserved.
Since their introduction in 1982,1 vinyl radical cycliza-
tions have proved to be extremely popular in organic
synthesis.2 Seminal studies by Beckwith and Stork’s
groups3,4 have shown that, under tin hydride mediated
reaction conditions, vinyl radical cyclization gives a
mixture of both 5-exo and 6-endo products 2a and 3a,
respectively (Scheme 1). More recently, Crich’s group
reported the preferential formation of 5-exo products
when the reaction was conducted in the presence of
PhSeSePh.5 The work by Beckwith and O’Shea3
revealed that formation of the more stable methylene-
cyclohexane derivative 3a is not due to a competing
6-endo [6-(p-exo)-endo–trig]6 cyclization (1a“6a), but it
is the result of a rapid rearrangement of the methylene-
cyclopentyl radical, 4a, via a reversible 3-exo–trig
cyclization (4a“5a“6a). As such, the 5-exo:6-endo
ratio is a function of tin hydride concentration with
higher concentrations favoring the kinetic 5-exo
product 2a.
On the other hand, it is well established that 5-alkyl
substituents retard radical cyclization of related 5-hex-
enyl radicals, so that 6-endo cyclization becomes an
effective competing reaction (see Scheme 2).2a,7
In this communication we describe experiments to
answer the question whether the formation of
methylenecyclohexane derivative 3b takes place via
‘direct’ 6-endo–trig cyclization (1b“6b) or via the inter-
mediates 4b, 5b and 6b (‘formal’ 6-endo mode5).2b,8
R
R
Bu3SnH
The yields of products 8, 9 and 10 (Scheme 3) obtained
when bromide, 7, was heated in toluene in the presence
of tributyltin hydride and 2,2%-azobisisobutyronitrile
(AIBN) at 80°C are given in Table 1.9 The same
products were also obtained by tin radical-mediated
cyclization10 of 1,6-enyne 11, followed by prot-
iodestannylation (4:2:1; AcOH, THF, H2O) (Table 1,
entries 2, 5, 8, 11 and 14).11 The above-mentioned
results have been compared with literature data for the
cyclization of the corresponding unsubstituted analog,
12.4
5-(π-exo)-exo-trig
2
4
R
R
+
5
1
5
R
1
R
Bu3SnH
R= H
6-(π-exo)-endo-trig
6
3
a) R= H, b) R) methyl
Stork and Mook reported cyclization of enyne 12 with
0.02 M Bu3SnH to give a 1:4 ratio of 5-exo and 6-endo
products 13 and 14, respectively (Table 1, entry 1).
Analogous cyclization of 7 and 11 (0.02 M Bu3SnH)
yielded methylenecyclohexane 912 as the sole observed
Scheme 1.
* Corresponding authors. Tel.: +34 91 5622900; fax: +34 91 5644853;
e-mail: anago@iqog.csic.es; clopez@iqog.csic.es
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00866-3

4998
A. M. Go´mez et al. / Tetrahedron Letters 43 (2002) 4997–5000
cyclization. This observation could be interpreted in
krel (exo)
krel (endo)
exo:endo
two different ways: (a) that formation of methylenecy-
clohexane 9 takes place, partially or completely, by
direct 6-endo–trig cyclization (i.e. 1b“6b) of 7 and 11,
or (b) that 3-exo cyclization of 4b is significantly faster
than 3-exo-cyclization of the unsubstituted analog 4a
(R=H) and therefore a pathway, 4b“5b“6b, could be
responsible for the formation of 9. In the first case the
rate of 6-endo–trig cyclization of radical 1b would be
faster than the rate of tin hydride transfer, and in the
latter the rate of 3-exo cyclization of 4b would prevail
over tin hydride transfer.
5
5
1.0
98:2
0.02
0.04
1
1
0.022
34:64
Scheme 2.
Br
In order to shed some light on this issue we prepared
iodide, 17 (Scheme 4), as a precursor of radical, 4b, by
iodination14 of hydroxy compound 16.15 Treatment of
iodide, 17, under the reaction conditions used in entries
14 and 15 (neat Bu3SnH), resulted in the exclusive
formation of methylenecyclopentane 8.16
Bu3SnH
+
+
E
E
E
E
E
E
E
E
9
10
8
7
This result shows that 3-exo radical cyclization of radi-
cal 4b is slower than hydride transfer in neat Bu3SnH,
and by corollary that formation of methylenecyclohex-
ane, 9 (Table 1, entries 14 and 15), had to be explained
by direct 6-endo–trig radical cyclization of radical 1b
(1b“6b). In addition, the fact that methylenecyclohex-
ane 9 prevailed over 8, when the cyclization was con-
ducted in neat Bu3SnH, seems to indicate that
6-endo–trig cyclization of 5-methyl vinyl radicals is
slightly faster than the corresponding 5-exo–trig ring
closure. Under lower tin hydride concentrations the
rearrangement 4b“5b“6b might also take place, there-
fore it could be assumed that methylenecyclohexane
formation in the cyclization of 1-vinyl-5-methyl-5-hex-
enyl radicals would take place by two different reaction
pathways, the formal 6-endo mode and a direct 6-endo–
trig mode.
1) Bu3SnH
8
+
9
+
10
2) AcOH, THF, H2O
E
E
11
SnBu3
SnBu3
SnBu3
Bu3SnH
ref. 4
+
+
E
E
E
E
E
15
E= CO2Me
E
E
E
13
14
12
Scheme 3.
product (Table 1, entries 2 and 3). Literature data for
cyclization of 12 with 0.25 M Bu3SnH showed a rever-
sal in the regiochemistry, and methylenecyclopentane
13 was obtained as the major isomer (Table 1, entry 4).4
Conversely, bromide 7, and enyne 11, cyclize under
similar, or even higher, tin hydride concentration (0.25
and 0.50 M) to give methylenecyclohexane 9 as the
major product13 (Table 1, entries 5 and 6). Radical
cyclization of 7 and 11 with 2.2 M Bu3SnH provided a
mixture of 9 and 8, with the former prevailing (Table 1,
entries 11 and 12), whereas cyclization of 12 with the
same Bu3SnH concentration had been reported to yield
exclusively the 5-exo product, 13 (Table 1, entry 10).
In summary, we have shown that in the vinyl radical
ring closure of 5-methyl hexenyl radicals, unlike vinyl
radical cyclization of 5-unsubstituted substrates, a
‘direct’ 6-endo–trig ring closure is responsible, to a
considerable extent, for the regiochemical outcome of
the reaction. It is, however, to be emphasized that the
regioselectivity control does not rely exclusively in the
presence of the methyl group, but that the tin hydride
concentration plays an essential role in the observed
regiochemistry. From a synthetic standpoint, these
results are of interest since they reveal that substituted
cyclohexanes can be efficiently prepared by cyclization
of 5-alkyl-vinyl radicals. This process is currently being
applied to the synthesis of carbasugars18,19 from carbo-
hydrates and the results will be reported in due course.
It was also reported that vinyl radical cyclization of 12
took place, even, in neat Bu3SnH leading to five-mem-
bered ring compound, 13, along with acyclic product,
15 (Table 1, entry 13).4 When a similar ring closure
reaction (neat Bu3SnH) was conducted on 5-methyl
analogs, 7 and 11, the six-membered ring product, 9,
was obtained as the major isomer (Table 1, entries 14
and 15).
I
OH
PPh3, I2
AIBN
Imidazole
neat Bu3SnH
E
E
E
E
E
E
These results (Table 1, entries 13–15) indicate that
radical cyclization of methyl substituted compounds 7
and 11 gives predominant 6-endo cyclization whereas
the hydrogen-substituted analog, 12, gives only 5-exo
17
16
8
E = CO2Me
Scheme 4.

A. M. Go´mez et al. / Tetrahedron Letters 43 (2002) 4997–5000
4999
Table 1. Radical cyclization of compounds 7, 11 and 12 under different tin hydride concentrations4,9,11
Entry
Substrate
Molar concentration
of substrate
Molar concentration of tributyltin Product ratio, Acyclic products Yield (%)a
hydride
5-exo:6-endo
(% yield)
1 (lit.)4
2
12
11
7
12
11
7
12
11
7
12
11
7
0.02
0.02
0.02
0.3
0.25
0.3
0.7
0.5
0.5
1.9
1.9
1.9
0.02
0.02
0.02
0.25
0.5
0.25
0.6
0.75
0.75
2.2
2.2
2.3
Neat Bu3SnH
Neat Bu3SnH
Neat Bu3SnH
20:80
0:100
0:100
80:20
16:84
15:85
95:5
19:81
28:72
100:0
30:70
38:62
100:0
43:57
45:55
–
–
72
65
3
4 (lit.)4
5
6
–
–
59
75
7 (lit.)4
–
8
9
–
58
63
14b
10 (lit.)4
11
21c
13b
16
54
65
12
13 (lit.)4
12
11
7
Neat Bu3SnH
Neat Bu3SnH
Neat Bu3SnH
14
15
28c
24b
72
58
a Based on isolated compounds.
b Yield of isolated 10.
c Compound 10 was not detected but instead a mixture of tin containing products were isolated. The structure shown below had been tentatively
assigned to one of the products.
SnBu3
E= CO2Me
E
E
Acknowledgements
to the ring formed; see: Crich, D.; Fortt, S. M. Tetra-
hedron Lett. 1987, 28, 2895.
7. Beckwith, A. L. J.; Ingold, K. U. In Rearrangements in
Ground and Excited States; de Mayo, P., Ed.; Academic
Press: New York, 1980.
This research was supported with funds from the Direc-
cio´n General de Ensen˜anza Superior (Grants No.
PB97-1244, PPQ2000-1330, and BQU2001-0582). We
are grateful to our colleague Dr. Jose Luis Chiara for
his help in the preparation of compound 16, which was
carried out in his laboratory. M.D.C. thanks the
Comunidad Auto´noma de Madrid for a scholarship.
C.U. thanks the Comunidad Auto´noma de Madrid for
a postdoctoral fellowship.
8. It has been shown, however, how vinyl radical cyclization
may be directed to give a high preference for 6-endo
cyclization by substituting the 5-position of the alkene
with radical stabilizing groups: (a) Urabe, H.; Kuwajima,
I. Tetrahedron Lett. 1986, 27, 1355; (b) Munt, S. P.;
Thomas, E. J. J. Chem. Soc., Chem. Commun. 1989, 480;
(c) Maguire, R. J.; Munt, S. P.; Thomas, E. J. J. Chem.
Soc., Perkin Trans. 1 1998, 2853.
9. General experimental procedure: A thoroughly degassed
(argon) soln of bromide 7 (85 mg, 0.28 mmol) in toluene
(appropriate concentration) was heated to 80°C under
argon. A soln of Bu3SnH (1.6 equiv.) and AIBN (0.1
equiv.) in toluene (appropriate concentration) was then
added and the reaction mixture was kept at that tempera-
ture over 6 h. After cooling, the organic solvent was
evaporated, the residue treated with CCl4 (0.2 mL) and a
dilute soln of iodine in ether until a faint color persisted.
The organic solvent was then removed in vacuo, the
residue taken up in ethyl acetate, washed with a satd aq.
soln of potassium fluoride, dried and evaporated. The
reaction mixture was then evaporated, the residue chro-
matographed (hexane/ethyl acetate; 98:2) and the prod-
ucts ratio determined by H NMR.
10. Stork, G.; Mook, R., Jr. J. Am. Chem. Soc. 1987, 109,
2829.
11. General experimental procedure: A thoroughly degassed
(argon) soln of enyne 11 (224 mg, 1.0 mmol) in toluene
was heated at 110°C under argon. A soln of Bu3SnH (1.5
References
1. Stork, G.; Baine, N. H. J. Am. Chem. Soc. 1982, 104,
2321.
2. (a) Giese, B.; Kopping, B.; Go¨bel, T.; Dickhaut, J.;
Thoma, G.; Kulicke, K. J.; Trach, F. Org. React. 1996,
48, 301; (b) Motherwell, W. B.; Crich, D. Free Radical
Reactions in Organic Synthesis; Academic Press: London,
1992; (c) Curran, D. P. Synthesis 1988, 417; (d) Curran,
D. P. Synthesis 1988, 489; (e) Jasperse, C. P.; Curran, D.
P.; Fevig, T. L. Chem. Rev. 1991, 91, 1237.
3. Beckwith, A. L. J.; O’Shea, D. M. Tetrahedron Lett.
1986, 27, 4525.
4. Stork, G.; Mook, R., Jr. Tetrahedron Lett. 1986, 27,
4529.
5. Crich, D.; Hwang, J.-T.; Liu, H. Tetrahedron Lett. 1996,
37, 3105.
1
6. (p-Exo)-vinyl radical cyclizations were defined as vinyl
radical cyclization in which the vinyl radical is exocyclic

5000
A. M. Go´mez et al. / Tetrahedron Letters 43 (2002) 4997–5000
equiv.) and AIBN (0.1 equiv.) in toluene (to reach the
14. Garegg, P. J.; Samuelsson, B. J. Chem. Soc., Perkin
final concentration given in Table 1) was then added and
the reaction mixture was kept at that temperature over 6
h. After cooling, the organic solvent was evaporated, the
residue was dissolved in a mixture of AcOH/THF/H2O
(4:2:1) (4 mL) and warmed at 85°C, with stirring for 1 h.
The reaction mixture was then evaporated and the
residue chromatographed (hexane–ethyl acetate; 7:3).
Trans. 1 1980, 2866.
15. Gansa¨uer, A.; Bluhm, H.; Pierobon, M. J. Am. Chem.
Soc. 1998, 120, 12849.
16. An analytical sample of 8 was prepared by deiodination
(Zn dust, aq. AcOH, 100°C)17 of 17: 1H NMR (200 MHz,
CDCl3) l (ppm): 4.88 (t, 1H, J=2.2 Hz), 4.79 (t, 1H,
J=2.2 Hz), 3.74 (s, 6H), 3.09 (t, 2H, J=2.2 Hz), 2.31 (s,
2H), 1.10 (s, 6H). Anal. Calcd for C12H18O4: C, 63.70; H,
8.02. Found: C, 63.96; H, 8.16.
17. Ito, M.; Kibayashi, C. Synthesis 1993, 137.
18. Suami, T.; Ogawa, S. Adv. Carbohydr. Chem. Biochem.
1990, 48, 21.
12. The spectral data obtained for compound 9 are in com-
plete agreement with known literature values. See: Ozaki,
S.; Horiguchi, I.; Matsushita, H.; Ohmori, H. Tetra-
hedron Lett. 1994, 35, 725.
1
13. The ratio of products (8:9) was determined by H NMR
19. (a) Go´mez, A. M.; Danelo´n, G. O.; Valverde, S.; Lo´pez,
J. C. J. Org. Chem. 1998, 63, 9626; (b) Go´mez, A. M.;
Danelo´n, G. O.; Moreno, E.; Valverde, S.; Lo´pez, J. C.
Chem. Commun. 1999, 175.
(CDCl3); 9: 4.74 (s, 2H, ꢀCH2
(dt, J=13.4, 2.0 Hz, 1H, CH2
Hz, 1H, ꢀCH
CHCH3).
6
), 3.71 (s, 3H, OCH3), 2.91
CHCH3); 8: 4.88 (t, J=2.2
6
6
2), 3.74 (s, 3H, OCH3), 3.09 (t, J=2.2, 1H,
6