2502
T. LIANG and S. KUWAHARA
3. Thus, the two-phase reaction mixture was stirred
for an additional three hours at room temperature
after the disappearance of 3. However, in this case
also, only a small amount of desired product 4 could
be detected by a 1H-NMR analysis of the crude
dehyde [(
2.65 mmol) in dichloromethane (15 ml) was added
dropwise a solution of CPBA (70 , 0.654 g,
2.65 mmol) in dichloromethane (10 ml) at ca
„10 C. The mixture was stirred at the same temper-
S)-1]. To a stirred solution of 3 (0.700 g,
m
z
.
9
product. Contrary to these reports, Hassner et al
.
ature for 30 min, poured into sat. aq. NaHCO3, and
extracted with chloroform. The extract was succes-
sively washed with water and brine, dried over
have reported the conversion of aldehyde-derived
silyl enol ethers to hydroxy acetals analogous to 4 by
treating with
m
CPBA in dichloromethane either in
MgSO4, and concentrated in vacuo to give 0.99 g
1
the presence or absence of solid sodium bicar-
bonate.7) The application of this non-aqueous proce-
dure to our substrate (3) led to the prompt formation
(86
z
) of 4. Judging from the H- NMR analysis of
the mixture, this product seemed to consist of all of
the four possible diastereomers. This mixture was
used directly for the next reaction. The major com-
ponent, however, could be puriˆed by silica gel
column chromatography (eluting with hexane-ethyl
of 4 in an 86
z
yield as a mixture of four
1
stereoisomers, as judged by its H-NMR spectrum.
These results imply that this transformation proceed-
ed much more e‹ciently under non-aqueous condi-
tions. With key intermediate 4 in hand, we subjected
hydroxy acetal 4 to PCC-oxidation conditions.8,9)
This oxidation, involving concurrent allylic rearran-
gement of the hydroxyl group, proceeded smoothly
acetate, 20:1). IR n
max cm„1: 3500 (m), 3090 (w), 3040
(w), 2970 (s), 2950 (s), 2875 (s), 1735 (vs), 1655 (m),
1580 (m), 1480 (m), 1440 (m), 1300 (s), 1270 (vs),
1180 (s), 1120 (s), 1080 (s), 950 (s), 900 (s), 850 (vs);
1H-NMR
d
: 0.06 (3H, s, SiCH3), 0.20 (3H, s, SiCH3),
to give a protected g-keto-a,b-unsaturated aldehyde
0.92 (9H, s, Si(CH3)3), 1.70–1.86 (4H, m, CH2CH2),
=
1.74 (3H, s, CCH3), 2.21 (1H, s, OH), 2.66–2.74
intermediate which, without puriˆcation, was
=
=
=
deprotected by treating with dilute hydro‰uoric acid
(1H, m,
5.77 (1H, dd,
dd, 10.3, 2.2 Hz, C
=
CCHC ), 4.76–4.80 (2H, m,
CH2),
10.3, 2.5 Hz, C CH), 5.88 (1H,
CH), 6.17 (1H, s, OCHO),
8.0 Hz, aromatic-5H), 7.56 (1H, ddd,
8.0, 2.0, 1.0 Hz, aromatic-4H), 7.95 (1H, ddd,
8, 1.5, 1.0 Hz, aromatic-6H), 8.04 (1H, dd, 2,
to aŠord the target molecule, (
S
)-1
s
[
a
]2D2+69.6
9
(
c
=
=
J
=
1.93, hexane)
possibility that the
readily migrate to the
ble robinal (5), (
t
. Although there was the undesirable
J
=
=
b
,
g
a
-double bond of (
S
)-1 might
7.41 (1H, t,
J
=
=
,
b
-position forming more sta-
J
J
S
)-1 could survive the acidic oxida-
J
=
tion conditions, while prolonged puriˆcation with a
silica gel column or standing at room temperature for
several weeks brought about isomerization to a sub-
1.5 Hz, aromatic-2H). To a stirred mixture of PCC
(1.08 g, 5.01 mmol) and pulverized 4A molecular
sieves (1.0 g) in dichloromethane (20 ml) was added
dropwise a solution of 4 (437 mg, 1.00 mmol) in
stantial extent. The same sequence of reactions was
22
applied to obtain (
hexane) from
(300 MHz) of (
chiral shift reagent, Eu(hfc)3, revealed the optical
purity of each enantiomer to be more than 95
R
)-1
s
[
a
]
„67.3
9
(
c
=
1.07,
9
dichloromethane (5 ml) at 0 C, and the mixture was
D
t
(
S
)-2. The 1H-NMR analyses
)-1 and ( )-1 in the presence of the
stirred for 4 hours at room temperature. The mixture
was diluted with ether and ˆltered through a pad of
Florisil}. The ˆltrate was concentrated in vacuo, and
the residue was dissolved in acetonitrile (10 ml). The
S
R
z.
Synthetic samples are now being investigated by Y.
Kuwahara (Kyoto University) to determine the abso-
lute conˆguration and to identify their biological
function
z
solution was mixed with a solution of 48 aq. HF
(0.5 ml) in water (5 ml), and stirred for 2 hours at
room temperature. The mixture was poured into sat.
aq. NaHCO3 and extracted with ether. The ethereal
solution was successively washed with water and
brine, dried over MgSO4, and concentrated in vacuo
.
Experimental
The residue was chromatographed over silica gel
(
S
)-Perillaldehyde was purchased from Aldrich
(15 g, petroleum ether-EtOAc, 10:1) to give 67.4 mg
22
=
1.93, hexane); IR
Chemical Co., and ( )-perillaldehyde was prepared
R
(41
z
) of (
S
)-1. [
a
]
+69.6
9
(
c
D
from (+)-limonene oxide according to the litera-
ture.10) IR spectra were measured as ˆlms by a Jasco
IR Report-100 spectrometer. 1H-NMR (300 MHz)
spectra were recorded with TMS as an internal stan-
dard in CDCl3 by a Varian Gemini 2000 spectrome-
ter. Optical rotation values were measured with a
Horiba Septa-300 polarimeter. Dichloromethane was
puriˆed by drying with P2O5 and then by distillation
from CaH2. Merck silica gel 60 (70–230 mesh) was
used for silica gel column chromatography.
nmax cm„1: 3070 (w), 2920 (s), 2820 (m), 2710 (w),
1690 (vs), 1450 (m), 1420 (m), 1285 (m), 1260 (m),
1110 (s), 1015 (m), 795 (m); 1H-NMR
d: 1.77 (3H, br
s, 4-CCH3), 2.11–2.19 (2H, m, 5-H2), 2.37–2.50 (1H,
=
18.8, 4.7, 1.1 Hz, 6-H),
m, 6-H), 2.67 (1H, ddt,
J
=
=
3.13 (1H, t,
CH), 5.00 (1H, qui,
dd,
W
J
7.8 Hz, 4-H), 4.77 (1H, br s, 4-C
1.5 Hz, 4-C CH), 6.58 (1H,
J
=
=
=
1.1, 2.3 Hz, 2-H), 9.79 (1H, s, CHO); MS
J
m z (relative intensity): 165 (13), 164 (M+, 100), 149
(57), 136 (20), 135 (31), 121 (15), 107 (26), 96 (54), 93
(19), 91 (20), 80 (11), 79 (26), 77 (17), 68 (75), 67 (48),
53 (20), 44 (19), 41 (18), 40 (36), 39 (34); HREIMS
(S)-4-Isopropenyl-3-oxo-1-cyclohexene-1-carbal-