5712
J . Org. Chem. 1996, 61, 5712-5713
Ch ir a l Bicyclic La cta m s. A New Stu d y on
F a cia l Alk yla tion
A. I. Meyers,* M. A. Seefeld, and B. A. Lefker1
Department of Chemistry, Colorado State University,
Fort Collins, Colorado 80523
F igu r e 1.
Received May 23, 1996
The chiral bicyclic lactams 1 have been shown to be
highly useful templates for the asymmetric construction
of compounds containing quaternary carbon centers.2 The
major stereochemical event surrounding this methodol-
ogy was the endo-alkylation of the enolate, 2 (Scheme
1). In a large number of cases, the preferred endo
approach was 10-50:1, yet the reasons for this were
vague at best.2,3 In contrast to other instances4 where
isomeric chiral lactams produced mainly exo-alkylated
products (eq 1) from their lithium enolate, lactam enolate
Sch em e 1
2 invariably furnished endo alkylated products, 3. The
synthetic utility of chiral lactams 3 was amply demon-
strated as they were transformed into a host of chiral
compounds, including 4,4-dialkylcyclopentenones 4 in
high enantiomeric purity.2,5
group larger than hydrogen in A might have the same
effect, and this led to consideration of enolate C. If this
could be successfully implemented then it would not only
expand the synthetic potential for generating the enan-
tiomeric chiral quaternary compounds of 4 but would also
go far in explaining the subtlety of the endo-alkylation
in A and the exo-alklylations in B.
Toward this end, we have constructed a series of
racemic bicyclic lactams 5 from simple 1,2-amino alco-
hols8 and levulinic acid, 7. It was not necessary at this
We now report that a slight remote modification in the
substitution of lactams 1 has a dramatic effect on the
exo/endo selectivity.6 First, it is important to note the
subtle difference between lactams 1 and that shown in
eq 1 (Figure 1). The former (A) contains an oxygen atom
at the bridgehead and a methylene group at the starred
position. The latter (B) has its oxygen at the starred
position and methylene group connected to the bridge-
head carbon. When models were constructed7 of both
enolates A and B, they suggested that the endo alkylation
pathway in B was somewhat inhibited by the psuedoaxial
hydrogen projecting down into the concave region. On
the other hand, enolate A has an oxygen in place of the
methylene and therefore only projects lone pairs. This
may not be sufficient to inhibit the endo entry in A, which
is indeed the major pathway.
juncture to prepare optically active lactams since the only
question we were concerned with was the diastereofacial
selectivity of 5. If the notion that properly placed
substitutents in 5 was a valid one, then chiral, nonrace-
mic lactams fulfilling this requirement would eventually
be evaluated.
As seen from Table 1, lactams containing gem-dimethyl
(entries 1 and 2), gem-diisopropyl (entries 3-5), and gem-
diphenyl (entries 6-8) all gave, after sequential meta-
lation-alkylation, 94-99% exo alkyl products. In the
first alkylation step (Table 1, entries 3 and 6) a 94:6 ratio
of exo/endo product 6 (R2 ) CH2Ph, R1 ) H) was obtained.
It was noted, however, that some epimerization occurred
due to difficulty in controlling the amount of excess base
present. This effect, as expected, was of no consequence
since in the second metalation-alkylation step, epimer-
ization of the doubly alkylated products 6 was not a
concern and the diastereoselectivity was >99% exofacial.
Furthermore, and of great significance, there was little
Due to our continued efforts to expand the synthetic
scope of the chiral lactams, we were most interested in
reversing the order of enolate facial alkylation and felt
that addition of a bulky substituent to A in the concave
region might impede alkylation (enolate C). Since hy-
drogen in B was sufficient to inhibit endo alkylation, the
trajectory of the alkyl halide toward the enolate must be
from a direction between the methylene group and the
starred position. Thus, it appeared reasonable that a
(1) Present address: Pfizer Central Research, Groton, CT.
(2) For a review on this subject, see: Romo, D.; Meyers, A. I.
Tetrahedron 1991, 47, 9503. A discussion on the endo/exo selectivity
is presented in this review (pp 9557-9564).
(3) Liotta, D.; Durkin, K. A. J . Am. Chem. Soc. 1990, 112, 8162.
(4) Thottathill, J . K.; Maniot, J . L.; Mueller, R. H.; Wong, M. K. Y.;
Kissick, T. P. J . Org. Chem. 1986, 51, 3140.
(5) Meyers, A. I.; Wanner, K. Th. Tetrahedron Lett. 1985, 26, 2047.
(6) For a similar dramatic effect in changing from endo to exo
alkylation see Roth, G. P.; Leonard, S. F.; Tong, L. J . Org. Chem. 1996,
61, XXXX. We thank Dr. Roth for sharing this information with us
prior to publication.
(7) In addition to models, some preliminary molecular modeling
studies were also performed that further indicated the steric interfer-
ence of an alkyl or aryl substituent to alkylation of the enolate carbon.
(8) Evans, D. A.; Carroll, G. L.; Truesdale, L. K. J . Org. Chem. 1974,
39, 914.
S0022-3263(96)00952-8 CCC: $12.00 © 1996 American Chemical Society