as the radical nature8 of the photostimulated reactions of 6
and 8 with Pb(OAc)4/I2. Moreover, these experiments also
helped to demonstrate the additional advantage of the direct
use of parent amides 6 and 8 in that, even if the amidyl
radicals were quenched by the direct H-abstraction from the
solvent,2g they could be regenerated by reaction with Pb-
(OAc)4/I2 and ultimately led to the clean formation of
cyclized products in high yield.
We then extended this methodology to the 7-exo vs 8-endo
cyclization system. We were delighted to find that regiospe-
cific 7-exo cyclization reactions of 16a-c proceeded smoothly
to afford the caprolactams 17a-c in satisfactory yields (eq
2). Moreover, exclusive eight-membered lactams 19a-f were
achieved in the reactions of amides 18a-f bearing an internal
vinylic halogen substituent (eq 3 and Figure 1). These are
Supporting Information). With an internal Cl-substituent, the
activation energy (Ea) for 6-exo cyclization jumped from 3.7
to 7.4 kcal/mol while the Ea for 7-endo cyclization stays
around 4.0 kcal/mol (entries 1 and 2, Table 2). A similar
but reversed pattern is computed for the terminal Cl-
substituted radical 23 (entry 3, Table 2). In both cases, the
activation energy difference is larger than 3 kcal/mol,
consistent with the experimental results.9 As a comparison,
the methyl-substituted radicals have lower activation energies
presumably because of the increase of electron density of
the CdC bond. In the meantime, the energy differences
become smaller. The calculated 1.6 kcal/mol difference in
activation energies for 23 (R ) Me) is also in excellent
agreement with the experimental data of 1.
It should be noted that in the few separated examples of
carbon-centered radicals in the literature the halogen-
substitution effect appears to be much less satisfactory.10
Then what accounts for this remarkable halogen-substitution
effect for amidyl radicals? The steric effect and the radical-
stabilizing effect of halogen atoms should play an important
role in controlling the regioselectivity. However, these are
not enough to explain the above results, as Br is of similar
size to Me while the radical stabilization energy of a Br atom
is even lower than that of a Me moiety.11 We believe that
the lone pair-lone pair electron repulsion between the N
Figure 1. ORTEP drawings of crystals 19e and 19f.
also the first examples of 7-exo and 8-endo cyclizations of
amidyl radicals. The only exception was that 1,6-H migration
predominated in the reactions of amides 20 leading to the
formation of δ-lactones 21 (82%, 69%, and 37% for X )
Cl, Br, and I, respectively). Again, without a halogen substi-
tution, amides such as 6-heptenamide gave only the iodolact-
onization products under the above experimental conditions.
The above halogen-substitution effect is therefore signifi-
cant not only in that it allows the radical processes to
overtake the ionic ones but also in that the regioselectivities
can be nicely directed by the vinylic halogen.
To gain more insight into the role of halogen substitution,
density functional calculations at the B3LYP/6-31G* level
were performed on the cyclization of radicals 22 and 23.
The results are summarized in Table 2 (also see the
(8) The electrophilic iodocyclization product of 8c (with NaHCO3/I2, rt)
was identified to be 6-iodomethyl-2,3,4,5-tetrahydropyridine-2-one in 37%
yield via a six-membered-ring closure, which in turn indicated that the
formation of 9 was not an ionic process.
(9) The activation energies for 1,5-H migration of radicals 22 (R ) Cl)
and 23 (R ) Cl) were also computed (B3LYP/6-31G*) to be 9.1 and 7.5
kcal/mol, respectively, much higher than the activation energies for the
preferred cyclization, in excellent agreement with the experimental results
in eqs 2 and 3.
(10) (a) Knapp, S.; Gibson, F. S.; Choe, Y. H. Tetrahedron Lett. 1990,
31, 5397. (b) Ishibashi, H.; Kobayashi, T.; Nakashima, S.; Tamura, O. J.
Org. Chem. 2000, 65, 9022. (c) Cassayre, J.; Gagosz, F.; Zard, S. Z. Angew.
Chem., Int. Ed. 2002, 41, 1783. (d) Kamimura, A.; Taguchi, Y. Tetrahedron
Lett. 2004, 45, 2335. (e) Padwa, A.; Rashatasakhon, P.; Ozdemir, A. D.;
Willis, J. J. Org. Chem. 2005, 70, 519. (f) Sharp, L. A.; Zard, S. Z. Org.
Lett. 2006, 8, 831.
Table 2. Calculated (B3LYP/6-31G*) Activation Energies
entry
radical
Ea(6-exo)b
Ea(7-endo)b
1
2
3
4
5
22/23 (R ) H)a
22 (R ) Cl)
23 (R ) Cl)
22 (R ) Me)
23 (R ) Me)
3.7
7.4
3.0
5.0
2.3
4.1
4.0
6.2
2.2
3.9
a Reference 2d. b In kcal/mol.
(11) Henry, D. J.; Parkinson, C. J.; Mayer, P. M.; Radom, L. J. Phys.
Chem. A 2001, 105, 6750.
Org. Lett., Vol. 8, No. 12, 2006
2649