Insight into the Rh-catalyzed cyclization of dissymmetrically racemic (±)-3,4-disubstituted 4-pentenal: Regio-, diastereo-, enantioselectivity, and kinetic resolution

Article abstract of DOI:10.1016/S0040-4020(00)01107-8

Rh-Catalyzed cyclization was applied for the kinetic resolution of dissymmetrical (±)-3-(1-methylvinyl)-4-phenylpent-4-enal. The cyclization by a chiral neutral Rh[(S)-BINAP]Cl afforded a mixture of 3,4-cis-4-methylcyclopentanone (>95% ee, 19% yield), 3,4-trans-4-methylcyclopentanone, and 3,4-cis-4-phenylcyclopentanone (>95% ee, 21% yield). The cyclization by a cationic Rh[(R)-BINAP]ClO₄ afforded a mixture of 3,4-trans-4-methylcyclopentanone (>95% ee, 36% yield), 3,4-cis-4-methyl-, and 3,4-trans-4-phenylcyclopentanone, accompanied with recovery of the starting material (-)-4-pentenal (54% ee, 34% yield), and by a cationic Rh[(S)-BINAP]ClO₄ afforded the corresponding enantiomers. The course of the cyclization and the stereochemistry of products are discussed based on the plausible acyl-hydride rhodium intermediates.

Full text of DOI:10.1016/S0040-4020(00)01107-8

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 1205±1211
Insight into the Rh-catalyzed cyclization of dissymmetrically
racemic (^)-3,4-disubstituted 4-pentenal: regio-, diastereo-,
enantioselectivity, and kinetic resolution
Masanori Imai, Masakazu Tanakap and Hiroshi Suemunep
Graduate School of Pharmaceutical Sciences, Kyushu University, Fukuoka 812-8582, Japan
Received 23 October 2000; accepted 20 November 2000
AbstractÐRh-Catalyzed cyclization was applied for the kinetic resolution of dissymmetrical (^)-3-(1-methylvinyl)-4-phenylpent-4-enal.
The cyclization by a chiral neutral Rh[(S)-BINAP]Cl afforded a mixture of 3,4-cis-4-methylcyclopentanone (.95% ee, 19% yield), 3,4-
trans-4-methylcyclopentanone, and 3,4-cis-4-phenylcyclopentanone (.95% ee, 21% yield). The cyclization by a cationic Rh[(R)-BINAP]-
ClO4 afforded a mixture of 3,4-trans-4-methylcyclopentanone (.95% ee, 36% yield), 3,4-cis-4-methyl-, and 3,4-trans-4-phenylcyclo-
pentanone, accompanied with recovery of the starting material (2)-4-pentenal (54% ee, 34% yield), and by a cationic Rh[(S)-
BINAP]ClO4 afforded the corresponding enantiomers. The course of the cyclization and the stereochemistry of products are discussed
based on the plausible acyl±hydride rhodium intermediates. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
alkylation with ethyl bromoacetate in 91% yield. The 1,3-
dicarbonyl function in 2 was converted to diene 3 in 30%
yield by treatment with the Nysted reagent {cyclodibromo-
di-m-methylene[m-(tetrahydrofuran)]trizinc}6 and TiCl4.
The 4-pentenal (^)-4 was prepared from the diene 3 in
88% yield by reduction with DIBAL-H at 2788C, as
shown in Scheme 1.
Asymmetric cyclization of 4-pentenals by Rh-complexes
was discovered independently by Bosnich1 and us.2 We
reported recently that Rh-catalyzed cyclization could be
applied to the symmetrical 3,4-disubstituted 4-pentenals
for the enantio- and diastereoselective reaction,3 and also
to the symmetrical 3,3,4-trisubstituted 4-pentenals for the
construction of a chiral quaternary carbon.4 In the case of
the symmetrical diene±aldehyde, very high enantiomeric
excess (.95% ee) and diastereomeric excess (.95% de)
of products were attained by using a neutral Rh[(R)- or
(S)-BINAP]Cl and a cationic Rh[(R)- or (S)-BINAP]-
ClO4.3b,5 Here we report the Rh-catalyzed cyclization of
dissymmetrically racemic diene±aldehyde (^)-4, as a part
of our ongoing research of Rh-catalyzed intramolecular
hydroacylation. In the case that racemic 4-pentenal (^)-4
is used as a substrate for the Rh-catalyzed cyclization, four
types of selectivity, that is to say, regioselectivity, diastereo-
selectivity, enantioselectivity, and kinetic resolution of
4-pentenal might exist in the cyclization.
Scheme 1.
2.2. Rh-catalyzed cyclization
The results of Rh-catalyzed cyclization are summarized in
Table 1. The cyclization by an achiral Rh(PPh3)3Cl afforded
a mixture of cis-5 and cis-6 in the ratio of 55 to 37 in 92%
yield, but no trans-products were given. Unfortunately,
these two constitutional isomers could not be separated by
column chromatography on silica gel. Therefore, the ratio of
isomers was determined by the intensity of the ole®n proton
peaks at d 5.38 (br s) and 5.00 (m) of cis-5 and 4.75 (m) and
4.54 (m) of cis-6 in the 1H NMR spectrum (Table 2). More-
over, the methyl proton signals appeared at d 0.70 (d,
2. Results and discussion
2.1. Preparation of 4-pentenals
Benzoylacetone 1 was ef®ciently converted to ester 2 by
Keywords: cyclization; rhodium; asymmetric reactions; kinetic resolution.
Corresponding authors. Tel.: 181-92-642-6604, fax: 181-92-642-6545;
e-mail: mtanaka@phar.kyushu-u.ac.jp; suemune@phar.kyushu-u.ac.jp
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)01107-8

1206
M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
Table 1. Asymmetric cyclization of dissymmetrical 3,4-disubstituted 4-pentenal (^)-4
Entry
Rh-complex (equiv.)
Reaction time (h)
Yields (%)
Recovery (%)
1
2
3
Rh(PPh3)3Cl (0.30)
Rh[(2)-DIOP]Cl (0.50)
Rh[(S)-BINAP]Cl (0.50)
24
5
72
55
43 (36% ee)
19 (.95% ee)
0
12
1
37
34 (64% ee)
21 (.95%
0
0
0
0
0
11a,b
ee)
0
0
4
5
Rh[(R)-BINAP]ClO4 (0.10)c
Rh[(S)-BINAP]ClO4 (0.10)
4
4
7
5
36 (.95% ee)
42 (.95% ee)
3
3
34d
43e
a
Acyl-hydride Rh-complex seemed to be formed and the yield of recovery was low.
[a]Dˆ21.08 (c 0.31, CHCl3).
b
c
d
Rh[(R)- or (S)-BINAP]ClO4 was prepared in situ from Rh[(NBD)BINAP]ClO4 by hydrogenation.
[a]Dˆ212.6 (c 1.0, CHCl3).
Jˆ7.6 Hz) in cis-5, whereas they appeared at d 1.38 (br s) in
at d 5.30 (br s) and 5.29 (br s) in the ratio of 68 to 32, and
that cyclized by the Rh[(S)-BINAP]Cl only indicated the
signal at d 5.30 (br s). Also, the H NMR spectrum of
(R,R)-2,3-butanediol acetal derived from (^)-cis-6 showed
the ole®n signals at d 4.64 (br s, 0.5H£2) and d 4.60, 4.57
(br s, each 0.5H) in the ratio of 1 to 1, but that of 6 cyclized
by Rh[(2)-DIOP]Cl showed the same signals in the ratio of
82 to 18, and that of cis-6 cyclized by the Rh[(S)-BINAP]Cl
only indicated the signal at d 4.64 (br s).
cis-6. The stereochemistry of the products was determined
unambiguously to be cis-5 and cis-6 by the NOESY 1H±1H
NMR spectrum, respectively. Correlation between the
methine proton signal at d 3.57 (m, 0.6H) of the
C(3)-position and the methine signal at d 2.50 (m, 0.6H)
of the C(4)-position in cis-5 was observed; also, correlation
between the methine signal at d 3.77 (br q, Jˆ7.0 Hz, 0.4H)
of C(3)-position and at d 3.21 (br q, Jˆ7.0 Hz, 0.4H) of the
C(4)-position in cis-6 was observed. The outcome of these
stereochemistries was consistent with the previous
results,2,3b namely, that the cyclization by the Rh(PPh3)3Cl
afforded 3,4-cis-products.
1
The cyclization by the cationic Rh[(R)- or (S)-BINAP]ClO4
afforded a mixture of cis-5, trans-5, and trans-6, but not cis-
6, as shown in entries 4 and 5. The ratio of products was
determined by the ole®n signals (Table 2). The dominant
stereoisomer was 3,4-trans-product 5, which was in accord
with the previous results3 that the cyclization by the cationic
Rh[BINAP]ClO4 afforded 3,4-trans-products. The relative
stereochemistry of the major product was unambiguously
determined to be trans-5 by the NOESY 1H±1H NMR
spectrum. Correlation between the methine proton signal
at d 2.89 (m, 0.78H) of the C(3)-position and the methyl
proton signals at d 1.15 (d, Jˆ7.6 Hz, 2.34H) of the C(4)-
position was observed. The enantiomeric excess of trans-5
was determined to be .95% ee by the 1H NMR spectra after
conversion of trans-5 into the corresponding (R,R)-2,3-
butanediol acetal. In entries 4 and 5, the starting material
4 was recovered, and the speci®c rotation showed
The cyclization by the chiral neutral Rh-complexes, such as
Rh[(2)-DIOP]Cl,5 Rh[(S)-BINAP]Cl afforded a mixture of
cis-5, trans-5, and cis-6, but not trans-6, as shown in entries
2 and 3 in Table 1. The isomers of the products could not be
separated by column chromatography on silica gel. The
ratio of products was determined by the intensity of the
1
ole®n proton signals in the H NMR spectra (Table 2).
The cis-products 5 and 6 were obtained dominantly by the
neutral Rh-complexes. The enantiomeric excess of cis-5 and
cis-6 was determined by the H NMR spectra after con-
1
version to the corresponding (R,R)-2,3-butanediol acetals.
The 1H NMR spectrum of (R,R)-2,3-butanediol acetal
derived from (^)-cis-5 showed the ole®n proton signals at
d 5.30 (br s) and 5.29 (br s) in the ratio of 1 to 1, while that
from cis-5 cyclized by Rh[(2)-DIOP]Cl showed the signals
[a]D ˆ212.6 (c 1.0, CHCl3, entry 4) and [a]27 ˆ112.7
31
D
1
(c 1.6, CHCl3, entry 5), respectively. The H NMR spectra
1
Table 2. Chemical shifts of cyclopentanones 5 and 6 in the H NMR spectra
Compound
Ole®nic-H
Methyl-H
5.38 (br s)
5.00 (m)
5.30 (m)
5.16 (m)
4.75 (m)
4.54 (m)
4.75 (m)
4.70 (m)
0.70 (d, Jˆ7.6 Hz)
1.15 (d, Jˆ7.6 Hz)
1.38 (br s)
1.69 (br s)

M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
1207
Figure 1. Plausible acyl±hydride intermediates for Rh(PPh3)3Cl-catalyzed cyclization.
of the (1)-MTPA ester derived from (^)-4 via the
corresponding primary alcohol, showed the ole®n signals
at d 4.76 (m) and 4.74 (m) in the ratio of 1 to 1, whereas
that derived from (2)-4 (entry 4) showed the signals in a
different ratio, and the enantiomeric excess of the recovered
(2)-4 was determined to be 54% ee (entry 4).
mediates (a, d), in which no steric repulsions exist. The
intermediates (b, c) which generate trans-products, seem
to be less stable due to the steric repulsions as indicated
by arrows in Fig. 1.
Fig. 2 shows the plausible acyl±hydride Rh-intermediates
with the (S)-BINAP ligand. Ellipsoids represent the phenyl
groups at phosphorus in the BINAP ligand, and imply
hindered regions. We assumed that two pseudo-axial phenyl
groups relative to the rhodium±phosphorus plane formed
the hindered regions at the vertical, and these regions
would strongly affect the stereoselection. Two pseudo-
equatorial phenyl groups may also form the hindered
regions at the horizontal, but these regions would exert
only minimal effect on the stereoselection. Also, we
assumed that the stereoselection by the neutral Rh[BI-
NAP]Cl is governed by the thermodynamically preferred
intermediate, and that by the cationic Rh[BINAP]ClO4 is
determined by the rate of reductive elimination from the
acyl±hydride intermediate.1,3,7 Taking these assumptions
into consideration, the cyclization by the neutral Rh[(S)-
BINAP]Cl could be explained by the thermodynamically
favorable intermediates. Four intermediates (f, g, i, l) of
the eight acyl±hydride intermediates (e±l) would produce
3,4-cis-products, and the intermediates (g, i) seem to be
more favorable than the intermediates (f, l) because no steric
repulsion exists in the intermediates (g, i). The intermediate
(g) from (3S)-4 generates (3R,4S)-cis-6, and the inter-
mediate (i) from (3R)-4 gives (3R,4S)-cis-5. Consequently,
both enantiomers (3S)- and (3R)-4-pentenals 4 could be
used for the cyclization by the neutral Rh[(S)-BINAP]Cl.
The results that the cyclization by the Rh[(S)-BINAP]Cl
2.3. Plausible mechanism for Rh-catalyzed cyclization
The mechanism for regioselection could not be clearly
explained. The difference in the electrostatic state and the
bulkiness between the phenyl and methyl substituents may
affect the course of cyclization.
The outcome of diastereoselectivity was in accord with the
previous results, that is to say, the cyclization by the neutral
Rh(PPh3)3Cl or Rh[BINAP]Cl complex afforded 3,4-cis-
cyclopentanone, and that by the cationic Rh[BINAP]ClO4
afforded 3,4-trans-cyclopentanone. These results may be
explained by the putative acyl±hydride rhodium inter-
mediates. We have already reported that the cyclization
by the neutral Rh-complex afforded cis-products by way
of the stable intermediate, in which no unfavorable steric
factors exist; on the other hand, the cyclization by the
cationic Rh[BINAP]ClO4 afforded trans-products by way
of the least stable intermediate of the putative acyl±hydride
intermediates because the rate of reductive elimination from
such intermediates might be faster than that of other inter-
mediates.3b,7
The cyclization of (^)-4 by the achiral Rh(PPh3)3Cl
afforded a mixture of cis-5 and cis-6 by way of the inter-

1208
M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
Figure 2. Plausible acyl±hydride intermediates for the chiral Rh-catalyzed cyclization.
afforded cis-5 (.95% ee) and cis-6 (.95% ee) and the
from the unfavorable intermediates seems to be faster than
recovered material 4 (entry 3) showed [a]27 ˆ21.088,
that from the favorable intermediates. Four intermediates (e,
h, j, k) would produce 3,4-trans-products. The rate of reduc-
tive elimination from the sterically unfavorable inter-
mediates (h, j), in which two steric repulsions exist, might
be faster than that from the intermediates (e, k), in which
one steric repulsion exists. The intermediate (h) from (3S)-4
generates (3R,4R)-trans-6, and the intermediate (j) from
D
which meant the recovered 4 was almost a racemate, accord
with the plausible mechanism.
The cyclization by the cationic Rh[(S)-BINAP]ClO4 affords
3,4-trans-products, which could be explained by the rate of
reductive elimination, and the rate of reductive elimination

M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
1209
(3R)-4 gives (3R,4R)-trans-5. In fact, the cyclization of (^)-
4 by the cationic Rh[(S)-BINAP]ClO4 afforded trans-5
(.95% ee) as a major product, but not trans-6 in moderate
yield. The ole®n bearing phenyl substituent seems to be less
reactive than the ole®n bearing an alkyl substituent in the
Rh-catalyzed cyclization.2b,d,3b The absolute con®guration
of trans-5 cyclized by Rh[(S)-BINAP]ClO4 might be
3R,4R, and that of the recovered (1)-4, which showed
33.0, 29.0, 14.0; FAB(1)HRMS calcd for C14H17O4
(M111) 249.1127, found 249.1129.
4.1.2. Ethyl 4-methyl-3-(1-phenylvinyl)pent-4-enoate (3).
A solution of 2 (10.0 g, 40.3 mmol) in THF (50 mL) was
added dropwise to the vigorous stirred suspension of Nysted
reagent5 (20% suspension in THF, 202 g, 88.7 mmol) in
THF (50 mL) at 2788C, and stirred for 15 min. Then,
TiCl4 (9.7 mL, 88.5 mmol) was added slowly to the stirred
mixture at 2788C, and then the whole was warmed to room
temperature, and stirred for 2 h. The mixture was diluted
with water, and extracted with ether. The etheral extract was
washed with 5% aqueous NaHCO3, brine, and dried over
MgSO4. After removal of the solvent, the oily residue was
puri®ed by column chromatography on silica gel (10%
EtOAc in hexane) to give 3 (2.9 g, 30%) as a colorless
[a]27 ˆ112.7, might be 3S.8
D
3. Conclusions
Rh-Catalyzed cyclization was applied for the kinetic resolu-
tion of racemic 3,4-disubstituted 4-pentenal (^)-4. The
cyclization by the achiral Rh(PPh3)3Cl afforded the cyclo-
pentanones cis-5 (55%) and cis-6 (37%), stereoselectively.
The cyclizations by the chiral Rh-complex proceeded enan-
tiospeci®cally to produce the optically active cyclopenta-
nones, that is to say, the cyclization of (^)-4 by the
neutral Rh[(S)-BINAP]Cl afforded cis-5 (.95% ee, 19%
yield) and cis-6 (.95% ee, 21% yield), and that by the
cationic Rh[(S)- or (R)-BINAP]ClO4 afforded trans-5
(.95% ee, 36±42% yields), accompanied with recovery
of the optically active starting material 4. The strategy
using the chiral Rh-complexes for the kinetic resolution
would be a practical and useful method to prepare optically
active cyclopentanones.
oil: IR (neat) 1730, 1620 cm21 1H NMR (270 MHz,
;
CDCl3) d 7.25±7.42 (m, 5H), 5.32 (br s, 1H), 5.10 (br s,
1H), 4.85 (m, 1H), 4.82 (br s, 1H), 4.09 (q, Jˆ7.3 Hz, 2H),
3.81 (t, Jˆ7.9 Hz, 1H), 2.64 (d, Jˆ7.9 Hz, 2H), 1.71 (br s,
3H), 1.21 (t, Jˆ7.3 Hz, 3H); FABMS m/z 245 (M111).
4.1.3. 3-(1-Methylvinyl)-4-phenylpent-4-enal (4). A solu-
tion of DIBAL-H in hexane (0.95 M, 4.0 mL, 3.77 mmol)
was added dropwise to a stirred solution of 3 (765 mg,
3.14 mmol) in toluene (7 mL) at 2788C, and the solution
was stirred for 1 h. The reaction was quenched with MeOH
(3 mL), and the whole was diluted with cold 1N HCl
(30 mL), extracted with EtOAc, washed with 5% aqueous
NaHCO3, brine, and dried over MgSO4. Removal of the
solvent afforded an oily residue, which was puri®ed by
column chromatography on silica gel (10% EtOAc in hex-
ane) to give 4 (550 mg, 88%) as a colorless oil: IR (neat)
4. Experimental
1
1730, 1630 cm21; H NMR (270 MHz, CDCl3) d 9.71 (t,
4.1. General methods
Jˆ2.3 Hz, 1H), 7.27±7.40 (m, 5H), 5.37 (br s, 1H), 5.10 (br
s, 1H), 4.91 (m, 1H), 4.82 (br s, 1H), 3.85 (t, Jˆ7.6 Hz, 1H),
2.71 (dd, Jˆ7.6, 2.3 Hz, 2H), 1.74 (br s, 3H); 13C NMR
(68 MHz, CDCl3) d 201.6, 148.9, 144.6, 141.5, 128.3,
127.6, 126.7, 114.2, 113.1, 46.1, 46.0, 20.8; FAB(1)HRMS
calcd for C14H17O1 (M111) 201.1279, found 201.1275.
Anhydrous THF was purchased from Kanto Chemical Co.,
and used as received. Benzene and CH2Cl2 were distilled
from P2O5. (R)- and (S)-BINAP were purchased from Kanto
Chemical Co. Inc. 1H NMR spectra were determined at 270
or 500 MHz. Infrared spectra were recorded on a JASCO A-
100 spectrometer. EIMS, FABMS, and HRMS spectra were
taken on a JEOL JMS 610H or SX102 spectrometer.
General procedures used for syntheses followed those of
the previous reports.3,4
4.1.4. Cyclization by the achiral Rh(PPh3)3Cl. A solution
of 4 (100 mg, 0.50 mmol) and Rh(PPh3)3Cl (139 mg,
0.15 mmol) in CH2Cl2 (16 mL) was stirred for 24 h under
Ar atmosphere. After removal of the solvent, the residue
was puri®ed by column chromatography on silica gel
(10% EtOAc in hexane) to leave a mixture of cis-5 and
cis-6 (92.3 mg, 92%, cis-5/cis-6ˆ6/4) as a colorless oil:
4.1.1. Ethyl 3-oxo-3-(phenylcarbonyl)pentanoate (2). A
solution of benzoylacetone (25.0 g, 154 mmol) in DMSO
(100 mL) was added dropwise to a stirred suspension of
NaH (60% in oil, 6.78 g, 170 mmol) in DMSO (150 mL)
at room temperature, and stirred for 30 min. The mixture
was cooled to 08C, and ethyl bromoacetate (18.9 mL,
170 mmol) was added dropwise. The whole was warmed
to room temperature, and stirred overnight, and then diluted
with saturated aqueous NH4Cl, extracted with ether, and
dried over MgSO4. After removal of the solvent, the residue
was puri®ed by column chromatography on silica gel (20%
EtOAc in hexane) to give 2 (34.8 g, 91%) as a colorless oil:
IR (neat) 1740 cm21 1H NMR (500 MHz, CDCl3) d
;
7.10±7.43 (m, 5H), 5.38 (br s, 0.6H), 5.00 (m, 0.6H), 4.75
(m, 0.4H), 4.54 (m, 0.4H), 3.77 (br q, Jˆ7.0 Hz, 0.4H), 3.57
(m, 0.6H), 3.21 (br q, Jˆ7.0 Hz, 0.4H), 2.35±2.79 (m, 4H),
2.10 (m, 0.6H), 1.38 (br s, 1.2H), 0.70 (d, Jˆ7.6 Hz, 1.8H);
EIMS m/z 200 (M1, 40), 170 (24), 156 (70), 130 (100), 115
(59), 77 (43); HRMS calcd for C14H16O1 (M1) 200.1201,
found 200.1201.
IR (neat) 1720, 1680, 1600 cm21 1H NMR (270 MHz,
;
4.1.5. Cyclization by the neutral Rh[(2)-DIOP]Cl. A
mixture of [Rh(cyclooctene)Cl]2 (90 mg, 0.12 mmol) and
bisphosphine (2)-DIOP (125 mg, 0.25 mmol) in CH2Cl2
(5 mL) was stirred at room temperature for 1 h, then a solu-
tion of 4 (100 mg, 0.50 mmol) in CH2Cl2 (2 mL) was added
dropwise to the stirred solution. After being stirred at room
CDCl3) d 8.01±8.05 (m, 2H), 7.49±7.67 (m, 3H), 5.01 (t,
Jˆ6.9 Hz, 1H), 4.14 (q, Jˆ7.3 Hz, 2H), 3.04 (dd, Jˆ6.9,
17.1 Hz, 1H), 2.96 (dd, Jˆ6.9, 17.1 Hz, 1H), 2.18 (s, 3H),
1.23 (t, Jˆ7.3 Hz, 3H); 13C NMR (68 MHz, CDCl3) d
201.8, 195.8, 171.2, 136.3, 133.8, 128.9, 128.7, 61.0, 57.7,

1210
M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
temperature for 5 h, the solution was concentrated in vacuo
to leave the residue, which was dissolved in ether (20 mL),
and the precipitated Rh-complex was ®ltered off. Removal
of the solvent gave the residue, which was puri®ed by
column chromatography on silica gel to afford a mixture
of cis-5, trans-5, and cis-6 (89 mg, 89%, cis-5/trans-5/
cis-6ˆ43/12/34) as a colorless oil.
Rh[(2)-DIOP]Cl showed the ole®n signals at d 5.30 (br s)
and 5.29 (br s) in the ratio of 68 to 32. Therefore, the
enantiomeric excess of cis-5 cyclized by Rh[(2)-DIOP]Cl
(entry 2 in Table 1) was determined to be 36% ee. The ole®n
proton signal of the (R,R)-butanediol acetal of cis-5 cyclized
by the neutral Rh[(S)-BINAP]Cl, was observed only at d
1
5.30 (br s) in the H NMR spectrum.
4.1.11. Determination of enantiomeric excess of cis-6.
The 500 MHz H NMR spectrum of the (R,R)-2,3-butane-
4.1.6. Cyclization by the neutral Rh[(S)-BINAP]Cl. The
cyclization by the neutral Rh[(S)-BINAP]Cl afforded a
mixture of cis-5, trans-5, and cis-6 (41%, cis-5/trans-5/
cis-6ˆ19/1/21) as a colorless oil. The starting material
4 was recovered in 11% yield. The speci®c rotation of
1
diol acetal of (^)-cis-6 showed the ole®n proton signals at d
4.64 (br s) and d 4.60, 4.57 (each, br s) in the ratio of 1 to 1,
while those from cis-6 cyclized by Rh[(2)-DIOP]Cl
showed the ole®n signals at d 4.64 (br s) and d 4.60, 4.57
(each, br s) in the ratio of 82 to 18, and those from cis-6
cyclized by the Rh[(S)-BINAP]Cl showed the signal at d
4.64 (br s), only.
recovered 4: [a]27 ˆ21.08 (c 0.31, CHCl3).
D
4.1.7. Cyclization by the cationic Rh[(R)-BINAP]ClO4. A
solution of [Rh(NBD)(R)-BINAP]ClO4 (34 mg, 0.038
mmol) in CH2Cl2 (5 mL) was stirred under H2 atmosphere
at room temperature for 2 h. Then, Ar gas was bubbled into
the solution for 15 min. This bright red solution of [Rh(R)-
BINAP]ClO4 was used for the cyclization without isolation.
A solution of 4 (150 mg, 0.75 mmol) in CH2Cl2 (5 mL) was
added dropwise to the stirred solution of [Rh(R)-BINAP]-
ClO4 under an Ar atmosphere. After being stirred at room
temperature for 4 h, the solution was concentrated in vacuo
to leave a residue. The residue was dissolved in ether
(20 mL), and the precipitated Rh-complex was ®ltered off.
After removal of ether, the residue was puri®ed by column
chromatography on silica gel. The fraction eluted with 15%
EtOAc in hexane afforded a mixture of 5 and 6 (69 mg,
46%, cis-5/trans-5/trans-6ˆ7/36/3) as a colorless oil and
eluted with 9% EtOAc in hexane gave recovered 4
4.1.12. Determination of enantiomeric excess of trans-5.
The 500 MHz H NMR spectrum of the (R,R)-2,3-butane-
1
diol acetal derived from trans-5 cyclized by the cationic
Rh[(R)-BINAP]ClO4 showed the methine proton signals at
d 2.58 (dt, Jˆ7.8, 11.4 Hz), while those cyclized by Rh[(S)-
BINAP]ClO4 showed the signals at d 2.65 (dt, Jˆ7.8,
11.4 Hz). Therefore, the enantiomeric excess of trans-5
(entry 5 and 6) was determined to be .95% ee.
4.1.13. Determination of enantiomeric excess of the
recovered starting material (4). LiAlH4 (4 mg,
0.10 mmol) was added portionwise to a stirred solution of
4 (10 mg, 0.05 mmol) in THF (4 mL) at 08C, and stirred at
room temperature for 3 h. The reaction was quenched with
diluted HCl, and then the whole was evaporated in vacuo.
The residue was dissolved in pyridine (5 mL) and four drops
of (1)-MTPACl was added. The solution was diluted with
brine, extracted with EtOAc, and dried over MgSO4. After
removal of the solvent, the residue was brie¯y puri®ed by
31
(51 mg, 34%): Recovered material 4: [a]D ˆ212.6 (c
1.00, CHCl3). A mixture of 5 and trans-6: IR (neat)
1
1740 cm21; H NMR (270 MHz, CDCl3) d 7.25±7.45 (m,
5H), 5.38 (br s, 0.15H), 5.30 (m, 0.78H), 5.16 (m, 0.78H),
5.00 (m, 0.15H), 4.75 (m, 0.07H), 4.70 (m, 0.07H), 3.67 (m,
0.15H), 2.89 (m, 0.78H), 2.10±2.65 (m, 4.29H), 1.83 (m,
0.78H), 1.69 (br s, 0.21H), 1.15 (d, Jˆ7.6 Hz, 2.34H), 0.70
(d, Jˆ7.6 Hz, 0.45H); HRMS calcd for C14H16O1 (M1)
200.1201, found 200.1204.
1
column chromatography on silica gel. The H NMR spec-
trum of (1)-MTPA ester derived from racemic (^)-4
showed the ole®n proton signal at d 4.76 (m) and 4.74
(m) in the ratio of 1 and 1, while those derived from the
recovered (2)-4 showed the signal at d 4.76 (m) and 4.74
(m) in the ratio of 23 and 77.
4.1.8. Cyclization by the cationic Rh[(S)-BINAP]ClO4.
The cyclization by the cationic Rh[(S)-BINAP]ClO4
afforded a mixture of 5 and 6 (50%, cis-5/trans-5/trans-
6ˆ5/42/3). The starting material 4 was recovered in 43%
yield.
References
1. (a) Barnhart, R. W.; Wang, X.; Noheda, P.; Bergens, S. H.;
Whelan, J.; Bosnich, B. J. Am. Chem. Soc. 1994, 116, 1821±
1830. (b) Barnhart, R. W.; McMorran, D. A.; Bosnich, B.
Chem. Commun. 1997, 589±590. (c) Bosnich, B. Acc. Chem.
Res. 1998, 31, 667±674.
4.1.9. (R,R)-2,3-Butanediol acetals of 5 and 6. A mixture
of cyclopentanone 5,6 (15 mg, 75 mmol), (R,R)-butanediol
(20 mg, 0.22 mmol), and p-TsOH±H2O (5 mg) in benzene
(20 mL) was re¯uxed for 3 h, ®xed with Dean±Stark
apparatus. After being cooled to room temperature, the solu-
tion was washed with 5% aqueous NaHCO3, brine, and
dried over MgSO4. After removal of the solvent, the residue
was brie¯y puri®ed by column chromatography on silica gel
to give the crude acetal.
2. (a) Sakai, K.; Ide, J.; Oda, O.; Nakamura, N. Tetrahedron Lett.
1972, 13, 1287±1290. (b) Taura, Y.; Tanaka, M.; Wu, X.-M.;
Funakoshi, K.; Sakai, K. Tetrahedron 1991, 47, 4879±4888.
(c) Sakai, K. J. Synth. Org. Chem. (Japan) 1993, 51, 733±
743. (d) Fujio, M.; Tanaka, M.; Wu, X. M.; Funakoshi, K.;
Sakai, K.; Suemune, H. Chem. Lett. 1998, 881±882.
3. (a) Wu, X.-M.; Funakoshi, K.; Sakai, K. Tetrahedron Lett.
1992, 33, 6331±6334. (b) Tanaka, M.; Imai, M.; Fujio, M.;
Sakamoto, E.; Takahashi, M.; Eto-Kato, Y.; Wu, X.-M.;
Funakoshi, K.; Sakai, K.; Suemune, H. J. Org. Chem. 2000,
65, 5806±5816.
4.1.10. Determination of enantiomeric excess of cis-5.
The 500 MHz H NMR spectrum of the (R,R)-2,3-butane-
1
diol acetal of (^)-cis-5 cyclized by the Rh(PPh3)3Cl showed
the ole®n proton signals at d 5.30 (br s) and 5.29 (br s) in the
ratio of 1 to 1, while those derived from cis-5 cyclized by

M. Imai et al. / Tetrahedron 57 (2001) 1205±1211
1211
4. (a) Takahashi, M.; Tanaka, M.; Sakamoto, E.; Imai, M.;
Funakoshi, K.; Sakai, K.; Suemune, H. Chem. Pharm. Bull.
2000, 48, 1822±1825. (b) Takahashi, M.; Tanaka, M.;
Sakamoto, E.; Imai, M.; Matsui, A.; Funakoshi, K.; Sakai, K.;
Suemune, H. Tetrahedron Lett. 2000, 41, 7879±7883.
6. Nysted reagent is commercially available from Aldrich Co.:
Matsubara, S.; Sugihara, M.; Utimoto, K. Synlett 1998, 313±
315.
7. Morrison, J. D. Asymmetric Synthesis, Vol. 5; Academic: New
York, 1985 (pp 46±66).
8. The absolute con®guration of trans-5 and (1)-4 was assumed
based on the plausible intermediates and the previous results of
Rh-catalyzed cyclizations.
5. Abbreviations: BINAP: 2,2(-Bis(diphenylphosphino)-1,1-
(-binaphthyl;
phosphinomethyl)-2,3-dimethyl-1,3-dioxolane.
(1)-DIOP:
(4S,5S)-(1)-4,5-Bis(diphenyl-