Communications
J . Org. Chem., Vol. 61, No. 17, 1996 5711
Ta ble 1. Mon o- a n d Bis-Alk yla tion of La cta m s 4a -c
entry
(compd) R1
R2
R3
basea
T (°C) exo:endob
1 (5a ) Ph Bn
Me s-BuLi
-80
-80
25
-80
-80
-80
-80
0
-80
-80
-80
-80
-80
0
>99:1c
96:4
92:8
2
3
4
5
6
Ph Me
Ph Me
Ph Bn
Ph Me
H
H
H
s-BuLi
LDA
s-BuLi
90:10
98:2
Bn s-BuLi
Ph allyl Me s-BuLi
>99:1c
>99:1
97:2
7 (5b) Me Me
H
H
H
H
H
s-BuLi
s-BuLi
F igu r e 2. ORTEP diagram of 5d .
8
9
10
11
Me Me
Me Me
Me Me
Me Me
s-BuLi/TMEDAd
s-BuLi/DMPUd
NaN(TMS)2
>98:2
>98:2
>98:2
>99:1
2:1
single-crystal X-ray structure of the product (Figure 2)11
serves as confirmation of the NMR assignment for exo
alkylation.
12 (5c) Me Bn
Me s-BuLi
13
14
15 (5d )
H
H
H
Me
Me
Bn
H
H
s-BuLi
s-BuLi
From a mechanistic standpoint, it is not immediately
obvious as to the nature of the high exo selectivity of the
alkylation reaction. Intuition would suggest a strong
steric effect, with the R-face of the enolate fully blocked
by either the bridging gem-dimethyl groups of the pinene
ring system or the axial methyl adjacent to the ring
oxygen, thus accounting for the high exo-selectivity even
at ambient temperatures. However, the poor monoalky-
lation selectivity of the hydrogen-substituted lactam 4c
does not fit this rationale, suggesting that there may
indeed be an electronic influence on the trajectory of
electrophilic attack.
1:1
Me s-BuLi
-80
>99:1c
a
b
All reactions were conducted at 0.1 M in THF. Diastereo-
meric ratio determined by 1H NMR at 270 MHz c Crude product
was recrystallized from hexanes; the minor diastereomer could not
d
be detected in the NMR of the crude product. An excess of 4 equiv
of the additive was used.
With this result, a series of experiments were con-
ducted in order to probe the preliminary scope of this
alkylation methodology. A summary of the results is
provided in Table 1. To our surprise, it was found that
substrate 4a will undergo exo alkylation with good
diastereoselectivity even when the reaction is conducted
at ambient temperature (Table 1, entry 3). Bis-alkylation
of substrates 4b and 4c with iodomethane and then
benzyl bromide also proceeded in an identical manner
(Table 1, entries 12 and 15). The exo selectivity was
verified via two methods, 1H NMR spectroscopy8 and
X-ray crystallography. Two-dimensional nuclear Over-
hauser spectroscopy9 (NOESY) spectra for 5b and c
(Table 1, entries 7 and 12) yield cross peaks from the
methyl hydrogens of R1 to the corresponding methyl
hydrogens of R2 for 5a and to the aromatic ring hydrogens
(2,6) for 5c (Table 1, entry 12). In addition, NOE cross
peaks from hydrogens of R1, R2, and or R3 to regiospecific
hydrogens on neighboring and proximal carbon atoms
confirm the unambiguous exo substitution. Similar
arguments stand true for compound 5d .
In summary, we have described a “Meyers-type” eno-
late-lactam system, derived from γ-keto acids, that
undergoes asymmetric alkylation in high enantiomeric
excess with unprecedented exo selectivity and where the
stereochemical induction can occur at unusually warm
temperatures. We believe that these observations extend
the synthetic utility of the “bicyclic lactam” approach to
enantiomerically pure compounds by providing access to
complementary enantiomers with a cost-effective tem-
plate and mild reaction conditions. Also, these observa-
tions are of theoretical interest generated by the opposite
sense of the selectivity.12 Work is currently in progress
to understand the mechanistic aspects for this selectivity
and to extend this methodology to the preparation of
more complex systems.
Su p p or tin g In for m a tion Ava ila ble: Experimental Pro-
cedures and copies of spectra are included (22 pages).
The addition of lithium-aggregate disrupting reagents
or use of the sodium enolate did not alter the course of
the reaction (Table 1, entries 9-11).
J O960852A
As reported for the [3.3.0] bicyclic lactams,10 the
hydrogen-substituted substrate 4c did not exhibit the
high diastereoselectivity on monoalkylation with io-
domethane (Table 1, entry 13). Subsequent re-enoliza-
tion and quenching with benzyl bromide furnished the
quaternary asymmetric center with high selectivity
(Table 1, entry 15, 5d ). It is worth noting that the
corresponding valinol-derived lactam described by Mey-
ers afforded only a 54% diastereomeric excess (endo
attack favored) for the same alkylation sequence.10 The
(11) Colorless crystals of the product 5d belong to the orthorhombic
crystal system, space group P212121, with a ) 7.764(1) Å, b ) 10.568-
(3) Å, c ) 23.395(7) Å. There are four molecules in the unit cell. The
final R factor is 6.8% and the final goodness-of-fit value is 2.02.
(12) A single related example has been reported in the literature to
give specific exo alkylation but was shown to readily epimerize to a
1:1 exo/endo mixture under thermodynamic conditions. Thottathil, J .
K.; Moniot, J . L.; Mueller, R. H.; Wong, M. K. Y.; Kissik, T. P. J . Org.
Chem. 1986, 51, 3140-3147. With our lactam system this is not the
case. Treatment of the methylated lactam 5b at ambient temperature
with sodium methoxide in methanol for 18 h only epimerized the newly
formed chiral center to a small extent, furnishing 5b with a 10:1 exo/
endo ratio. This is essentially identical to entry 3 in Table 1.
(8) NMR spectral assignments were performed using 1D and 2D
phase-sensitive DQ filtered COSY and NOESY experiments (mixing
time of 500 ms at 27 °C in CDCl3 at 500 or 600 MHz (approximately
3-5 mg/mL)).
(9) J eener, J .; Meier, B. H.; Bachmann, P.; Ernst R. R. J . Chem.
Phys. 1979, 71, 4546-4553.
(10) Lefker, B. A. Ph.D. Dissertation, Colorado State University,
Fort Collins CO, 1988.