An intramolecular Mitsunobu reaction of 4 was performed
with diethyl azodicarboxylate (DEAD) and PPh3 under high
dilution conditions (0.01 M), providing desired cyclic amide
1a in good yield (73%). It should be noted that only a
negligible amount of dimerized product (<1%) was formed
in this reaction.
Fortunately, amide 1a afforded a crystal suitable for X-ray
analysis. As shown in Figure 1, the C3-C4 and C7-C8
Figure 2. CD spectra and specific rotation of 1a.
starting from enantioenriched 1a. At the outset, we examined
a reductive detosylation of (+)-1a (>98% ee) using lithium
naphthalenide (LiNaph). The reaction successfully has
provided secondary amine (-)-1b in good yield without any
racemization.12 The obtained amine readily gave a variety
of ammonium salts upon treatment with carboxylic acids.
Ammonium salt (+)-6 prepared from amine (-)-1b and (S)-
R-methoxyphenylacetic acid [(S)-5] gave a crystal suitable
for X-ray analysis, which has led to the determination of
the absolute configuration of (-)-1b and its precursor (+)-
1a as S (Scheme 2) (Figure 3).6,12
Figure 1. Molecular structure of 1a.
olefinic moieties form chiral planes in the solid state.6,7
Moreover, the 1H NMR analysis of amide 1a in CDCl3 at
ambient temperature shows two sets of nonequivalent R-N-
allylic geminal protons (at C2 and C9), which suggests that
the above-mentioned chiral planes are maintained even in
solution.8 Analytical and semipreparative-scale HPLC using
a chiral stationary column equipped with a CD spectropo-
larimeter successfully separated both enantiomers of 1a as
shown in Figure 2.9,10
Scheme 2. Synthesis of Amine 1b and Its Salt 6
Enantiopurity of 1a was unchanged in the solid state
(crystal) at ambient temperature for at least 2 months. It
should be noted that 1a only racemizes at a very slow rate
even in solution: the half-life of optical activity of 1a in
hexane at 25 °C was estimated as 203 days.11 Encouraged
by these results, we next performed several transformation
High crystallinity of the salt 6 suggested a possibility
for optical resolution of amine 1b by fractional crystal-
lization. In fact, diastereomerically enriched (+)-6 (93%
dr, 42% yield) deposited preferentially from ether solu-
tion of a 1:1 mixture of rac-1b and (S)-5, and enantio-
enriched 1b (62% ee, R)12 was recovered from mother
liquor. Further recrystallization of (+)-6, followed by tosy-
lation provided an enantiomerically pure amide (S)-1a
(eq 1).
(3) We used a slightly modified procedure of the one described by:
Marshall, J. A.; Lebreton, J. J. Org. Chem. 1988, 53, 4108-4112.
(4) For reviews, see: (a) Mitsunobu, O. Synthesis, 1981, 1-28. (b)
Hughes, D. L. Org. React. 1992, 42, 335-656.
(5) Henry, J. R.; Marcin, L. R.; McIntosh, M. C.; Scola, P. M.; Harris,
G. D., Jr.; Weinreb, S. M. Tetrahedron Lett. 1989, 30, 5709-5712.
(6) The structures of rac-1a, (+)-6, and rac-7 were determined by X-ray
crystallography; see the Supporting Information. In Figures 1 and 3,
hydrogen atoms have been omitted for clarity, except the hydrogen attached
to the nitrogen.
(7) Note that the sum of the nitrogen’s bond angles is 348.2°, which
means that the nitrogen center of 1a is chiral in the solid state.
(8) Selected 1H NMR data of 1a (300 MHz, CDCl3): δ 3.00 (d, J )
10.2 Hz, 1H; C2-H), 3.01 (dd, J ) 14.1, 4.2 Hz, 1H; C9-H), 3.86 (dd, J )
14.1, 11.7 Hz, 1H; C9-H), 4.25 (d, J ) 10.2 Hz, 1H; C2-H).
(9) Analytical HPLC: CHIRALCEL OD-H (4.6 × 250 mm), hexane/
EtOH ) 1/1, 0.5 mL/min, UV 254 nm, rt, tR ) 9.9 min for (-)-(R)-isomer,
13.1 min for (+)-(S)-isomer. Preparative HPLC: CHIRALCEL OD-H (20
× 250 mm), hexane/EtOH ) 1/1, 3.0 mL/min, UV 254 nm, rt, tR ) 30.2
min for (-)-(R)-isomer, 41.0 min for (+)-(S)-isomer.
(10) CD spectroscopy was measured by a stopped-flow procedure.
(11) The half-life was calculated by the chiral HPLC measurement of
enantiopurity; enantiopurity of 1a decreased to 93% ee from >98% ee within
21 days in hexane at 25 °C.
Figure 3. Molecular structure of (+)-6 derived from (+)-1a.
964
Org. Lett., Vol. 8, No. 5, 2006
Tomooka, Katsuhiko
