Stereoselective Thio-Michael/Aldol Tandem Reaction to α,β-Unsaturated Esters

Article abstract of DOI:10.1021/jo990548s

A mixture of lithium thiophenolate, α,β-unsaturated ester, and aldehyde in CH₂Cl₂ afforded Michael/ aldol tandem adducts, β-hydroxy-α-(1-phenylthioalkyl) esters, in moderate to good yields with a high syn-aldol selectivity. The reaction proceeded effectively when CH₂Cl₂ or ether was used as a solvent. The countercation of thiolate proved important; lithium cation provided the best results. The stereoselectivity and yield of the adducts greatly depended on the steric size of the ester group. Lithium selenolate, generated in situ by treatment of diphenyldiselenide with methyllithium, brought similar results, whereas alkoxide was too inert for the reaction. Application of the reaction to methacrylate provided a useful method to form a quaternary carbon center on a similar stereoselective level. In the reaction with crotonates, on the other hand, the anti-Michael selectivity dominated over the syn-aldol selectivity. The tandem adduct was useful for the stereoselective preparation of trisubstituted tetrahydrofuran via radical cyclization, whose configuration was found to be 2,3-trans-3,5-trans.

Full text of DOI:10.1021/jo990548s

J . Org. Chem. 1999, 64, 6353-6360
6353
Ster eoselective Th io-Mich a el/Ald ol Ta n d em Rea ction to
r,â-Un sa tu r a ted Ester s
Akio Kamimura,* Hiromasa Mitsudera, Shigeru Asano, Seiji Kidera, and Akikazu Kakehi†
Department of Applied Chemistry, Faculty of Engineering, Yamaguchi University, Ube 755-8611 J apan,
and Department of Material Engineering, Faculty of Engineering, Shinshu University,
Nagano 380-8553 J apan
Received March 29, 1999
A mixture of lithium thiophenolate, R,â-unsaturated ester, and aldehyde in CH2Cl2 afforded Michael/
aldol tandem adducts, â-hydroxy-R-(1-phenylthioalkyl) esters, in moderate to good yields with a
high syn-aldol selectivity. The reaction proceeded effectively when CH2Cl2 or ether was used as a
solvent. The countercation of thiolate proved important; lithium cation provided the best results.
The stereoselectivity and yield of the adducts greatly depended on the steric size of the ester group.
Lithium selenolate, generated in situ by treatment of diphenyldiselenide with methyllithium,
brought similar results, whereas alkoxide was too inert for the reaction. Application of the reaction
to methacrylate provided a useful method to form a quaternary carbon center on a similar
stereoselective level. In the reaction with crotonates, on the other hand, the anti-Michael selectivity
dominated over the syn-aldol selectivity. The tandem adduct was useful for the stereoselective
preparation of trisubstituted tetrahydrofuran via radical cyclization, whose configuration was found
to be 2,3-trans-3,5-trans.
In tr od u ction
chemistry of the thio-tandem process with R,â-unsatur-
ated esters, despite the few reports of the reaction with
R,â-unsaturated ketones,7 had remained unestablished:
neither had the reaction conditions been opitimized nor
had the stereochemical course of the reaction been
known.8 Recently, we have demonstrated that the tan-
dem strategy successfully works for acrylates with the
aldehydes in CH2Cl2 and that a high syn-aldol selectivity
is achieved with tert-butyl acrylate.9,10 This research
reports in full detail our latest results on the tandem
process, discussing a plausible pathway of the reaction
and its stereochemical course. We will also demonstrate
a stereoselective preparation of trisubstituted tetrahy-
drofuran derivatives from the tandem adduct as an
extension of the methodology.
The tandem or domino reaction has recently been of
interest for organic synthesis because it offers a conve-
nient and economical way to prepare desired organic
molecules.1 Additionally, combining more than one reac-
tion in one pot usually provides a good solution, a solution
that will give a better yield and a chance for reactive but
not-easy-to-generate intermediates to be used in a syn-
thetic sequence. The Michael addition is one of the most
useful organic reactions to construct carbon skeletons.2
Thiolate and its analogues are known as good nucleo-
philes for the reaction to give Michael adducts in good
yields. The reaction begins with the nucleophilic attack
of thiolate on the â-carbon of a Michael acceptor, gener-
ating an enolate intermediate. If the active enolate in-
termediate is captured with not a proton but an aldehyde,
then a tandem Michael/aldol process is achieved.3-5 The
thio-functional group introduced here serves as a precur-
sor of other functional groups and/or acts as a good ac-
tivator for a further carbon-carbon bond-forming reac-
tion.6 When the present investigation was launched, the
Resu lts a n d Discu ssion
Th e Ta n d em Mich a el/Ald ol Rea ction to Acr yla te
Ester s. The present reaction procedure is outlined in
Scheme 1. First, methyl acrylate was chosen for the
Michael acceptor. The results are summarized in
Table 1.
† Shinshu University.
To a heterogeneous mixture of lithium thiolate in
(1) For review for tandem reaction, see: Posner, G. H. Chem. Rev.
1986, 86, 831. Tietze, L. F.; Beifuss, U. Angew. Chem., Int. Ed. Engl.
1993, 32, 131. Bunce, R. A. Tetrahedron 1995, 48, 13103. Tietze, L. F.
Chem. Rev. 1996, 96, 115.
(2) J ung, M. E. In Comprehensive Organic synthesis; Trost, B. M.,
Ed.; Pergamon Press: Oxford, 1991; Vol. 4, pp 1-67. Perlmutter, P.
Conjugate Addition Reactions in Organic Synthesis; Pergamon Press:
Oxford, 1992.
CH2Cl211 were added methyl acrylate and benzaldehyde.
(6) For example: Simpkins, N. S. Sulphones in Organic Synthesis;
Pergamon Press: Oxford, 1993.
(7) (a) Barrett, A. G. M.; Kamimura, A. J . Chem. Soc., Chem.
Commun. 1995, 1755. (b) Yura, T.; Iwasawa N.; Mukaiyama, T. Chem.
Lett. 1986, 187. (c) Suzuki, M.; Kawagishi, T.; Noyori, R. Tetrahedron
Lett. 1981, 22, 1809.
(8) Itoh, A.; Ozawa, S.; Oshima, K.; Nozaki, H. Tetrahedron Lett.
1980, 21, 361; Levin, J . I. Synth. Commun. 1992, 961.
(9) Preliminary communication on this work: Kamimura, A.; Mit-
sudera, H.; Asano, S.; Kakehi, A.; Noguchi, M. Chem. Commun. 1998,
1095.
(10) Very recently, a related work has been published: Ono, M.;
Nishimura, K.; Nagaoka, Y.; Tomioka, K. Tetrahedron Lett. 1999, 40,
1509.
(3) J ansen, J . F. G. A.; Feringa, B. L. Tetrahedron Lett. 1991, 32,
3239.
(4) Davies, S. G.; Ichihara, O. J . Synth. Org. Chem. J pn. 1997, 55,
26 and references therein. Davies, S. G.; Fenwick, D. R. J . Chem. Soc.,
Chem. Commun. 1997, 565. Tsukada, N.; Shimada, T.; Gyoung, Y. S.;
Asao, N.; Yamamoto, Y. J . Org. Chem. 1995, 60, 143. Yamamoto, Y.;
Asao, N.; Yuehara, T. J . Am. Chem. Soc. 1992, 114, 5427. Hosomi, A.;
Yanagi, T.; Hojo, M. Tetrahedron Lett. 1991, 32, 2371. Asao, N.;
Uyehara, T.; Yamamoto, Y. Tetrahedron 1990, 46, 4563.
(5) Fleming, I.; Kilburn, J . D. J . Chem. Soc., Chem. Commun. 1986,
305. 1198; Fleming, I.; Sarkar, A. K. J . Chem. Soc., Chem. Commun.
1986, 1199.
(11) Lithium thiolate was generated by treatemnt of thiophenol in
CH2Cl2 with BuLi at -78 °C, where no reaction between CH2Cl2 and
BuLi was observed except a clean formation of lithium thiophenolate.
10.1021/jo990548s CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/04/1999

6354 J . Org. Chem., Vol. 64, No. 17, 1999
Kamimura et al.
Sch em e 1
1k ) 90:10 (Scheme 2). This result thus indicates that
in the reaction the aromatic aldehydes are more reactive
than the aliphatic aldehydes.
We further applied the present procedure to selenolates
or alkoxides since analogues of thiolates, selenolates for
one, work as good nucleophiles in the Michael addition
reaction (Scheme 3). The results are summarized in
Table 2.
The reaction mixture became homogeneous and was
maintained at -78 °C for 1 h; desired tandem adduct 1a
was obtained in 31% yield along with the simple Michael
adduct of the thiol and the acrylate (Table 1, entry 1).
The reversed-phase HPLC analysis indicated that the
diastereomeric ratio of 1a was 71/29, with syn-1a found
to be the major isomer. The two diastereomers of 1a were
separated by flash chromatography. Neither diastere-
omer of diastereomerically pure 1a epimerized when they
were each exposed to lithium thiolate in CH2Cl2. This
initial result encouraged an attempt at optimizing the
reaction conditions so as to improve the yield and the
stereoselectivity. When the reaction conditions were set
at -50 °C for 7 h, the yield of 1a increased to 62% but
the syn selectivity remained at a similar level (Table 1,
entry 2). Cyclohexyl acrylate ensured the formation of
1c with a slightly better syn selectivity (79/21) but with
a moderate yield (Table 1, entry 4). Both the yield and
the syn selectivity were dramatically improved when tert-
butyl acrylate was used in the reaction; the desired
tandem adduct 1d was isolated in 80% yield with a 92/8
syn/anti ratio (Table 1, entry 5). This stereoselectivity is
much higher than in the analogous aldol reaction of the
ester enolate generated from tert-butyl propionate with
LDA, the reported ratio being almost 1:1.12 2,6-Dimeth-
ylphenyl acrylate, an ester residue much more bulky than
tert-butyl acrylate, underwent no formation of the tandem
adduct (Table 1, entry 6). Consequently, we concluded
that tert-butyl ester was the best choice for the ester part
of the reaction.
Changes in the countercation varied the yields and the
selectivities. Sodium thiolate, for example, afforded 1d
in a moderate yield (51%), but potassium thiolate was
ineffective for the formation of 1d (Table 1, entries 7 and
8). The presence of magnesium ions produced a mixture
of 1d in a low yield (Table 1, entry 9).13 A variety of
solvents were also examined for the reaction: ether was
similar to CH2Cl2 in the results (Table 1, entries 5 and
11), while THF and C2H5CN were found useless (Table
1, entries 10 and 12). Thus, lithium ions in either
CH2Cl2 or ether proved the best combination for a smooth
reaction. These conditions were further applied to reac-
tions of other aldehydes.
Our investigation needed lithium selenophenolate.
Although it is usually generated through deprotonation
of selenophenol,14 it was instead generated directly from
methyllithium and diphenyldiselenide.15 To the lithium
selenolate in ether, benzaldehyde first and then tert-butyl
acrylate were added at -50 °C, and the mixture gave
corresponding tandem adduct 2a in 97% yield (Table 2,
entry 1). The syn/anti ratio of 2a turned out to be 87/13,
almost equivalent to that of the thiolate reaction. Other
aromatic aldehydes also underwent the smooth formation
of 2 in good yields (Table 2, entries 2-5). The adduct of
R,â-unsaturated aldehyde was isolated in a good yield
but was a 1:1 mixture of the two diastereomers (Table 2,
entry 6). The reaction with aliphatic aldehyde resulted
in the formation of tandem adduct 2f in a moderate yield
(Table 2, entry 7). Phenoxide and benzyl alkoxide anion
neither caused the Michael addition nor formed oxy-
tandem adducts 3 under the present reaction conditions
(Table 2, entries 8 and 9).
Th e Ta n d em Mich a el/Ald ol Rea ction to Meth -
a cr yla te Ester s. The present procedure was extended
to methacryl esters providing a quaternary carbon center
(Scheme 4 and Table 3).
The same sequence for methacrylate at -15 °C, for
instance, afforded tandem adduct 4a in 61% yield at a
syn/anti ratio of 87/13 (Table 3, entry 1). The reaction
under the conditions of -78 °C to room temperature
slightly improved the yield but lowered the syn selectivity
(Table 3, entry 2). The moderate reactivity of methacry-
late ester toward the Michael addition required a higher
reaction temperature for the smooth reaction progress.
X-ray crystallographic analysis revealed the stereochem-
istry of the adduct to be 2R*,3S*, indicating that the
reaction with methacrylate preferred to proceed through
a stereochemical course similar to that of the reaction
with acrylates (vide infra). This procedure proved com-
patible with other aromatic aldehydes, the reaction of
which gave tandem adducts 4 in good yields in a syn-
selective manner (Table 3, entries 3 and 4). Thus, the
present method provides a useful preparation of quater-
nary carboncenters in a diastereoselective manner.16
Selenium analogue 5 was also prepared in the reaction
The reaction with aromatic and R,â-unsaturated alde-
hydes gave tandem adducts in good yields with a high
syn selectivity (Table 1, entries 13-16). On the other
hand, the yield of the tandem adducts with aliphatic
aldehydes remained at a moderate level, and mixtures
were isolated of the respective two diastereomers of
1j-l in a ratio of ca. 3:1 (Table 1, entries 17-19). The
R-branched aldehyde, such as isobutyraldehyde, was
inert in the reaction (Table 1, entry 20). A competitive
experiment between hexylaldehyde and benzaldehyde
resulted in the selective formation of 1d at a ratio of 1d /
(14) For generation of selenolate anion, see: Paulmier, C. Selenium
Reagents and Intermediates in Organic Synthesis; Pergamon: Oxford,
1986; pp 25-27 and references therein. Liotta, D. Acc. Chem. Res. 1984,
17, 28. Clive, D. L. J . Tetrahedron 1978, 34, 1049. Reich, H. Acc. Chem.
Res. 1979, 12, 22.
(15) Due to the high toxicity and terrible smell of selenophenol, we
wished to avoid to use this substance and looked for a convenient
alternative for our purpose. After several trials, we finally succeeded
in finding a useful and safer method, in which lithium selenophenolate
was generated by treatment of diphenyldiselenide with an equimolar
amount of methyllithium in ether at the room temperature (see
Experimental Section). Although half of the organoselenium source is
wasted in forming methyl phenyl selenide, the most advantageous
point of the method is that the generation of selenolate is achieved in
one step from commercially available diphenyldiselenide, and this also
minimizes unavoidable spread of the toxic selenium reagent during
the experimental work.
(12) Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, M.
C.; Sohn, J . E.; Lampe, J . J . Org. Chem. 1980, 45, 1066.
(13) Shono, T.; Matsumura, Y.; Kashimura, S.; Hatanaka, K. J . Am.
Chem. Soc. 1979, 101, 4752.
(16) For review for stereoselective construction of quaternary ste-
reogenic center, see: Fuji, K. Chem. Rev. 1993, 93, 2037.

Thio-Michael/Aldol Tandem Reactions
J . Org. Chem., Vol. 64, No. 17, 1999 6355
Ta ble 1. Mich a el/Ald ol Ta n d em Ad d ition to Acr yla tes u n d er Va r iou s Con d ition s
entry
R
R1
M+
Li
solvent
T (°C)
time (h)
1; yield (%)a
syn/antib
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Me
Me
Et
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
THF
-78
1
7
7
7
7
15
15
15
15
15
15
15
7
1a ; 31
1a ; 62
1b; 64
1c; 62
1d ; 80
1e; 0
1d ; 51
1d ; 0
1d ; 27
1d ; 33
1d ; 86
1d ; 0
1f; 91
1g; 71
1h ; 52
1i; 66
1j; 31
1k ; 65
1l; 66
1m ; 0
71/29
71/29
66/34
79/21
92/8
Li
Li
Li
Li
Li
Na
K
MgBr
Li
Li
Li
Li
Li
Li
Li
Li
Li
Li
Li
-50
-50
c-Hex
t-Bu
2,6-DMPc
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
t-Bu
-50
-50
-78 to rt
-50
86/14
-50
-50
63/37
71/29
87/13
-50
ether
-50
C2H5CN
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
-50
1-Napd
-50
88/12
88/12
81/19
85/15
76/24
72/28
76/24
p-ClC6H4
PhCHdCH-e
C3H7CHdCH-e
Me
-50
7
7
7
15
7
-50
-50
-78 to rt
-50
C5H11
C9H19
-78 to rt
-78 to rt
15
15
iPr
a
b
d
Isolated yield. Determined by HPLC analyses. c 2,6-DMP ) 2,6-dimethylphenyl. 1-Nap ) 1-naphthyl. e E isomer was used.
Sch em e 2
Determination of the other diastereomers will be dis-
cussed in the following section.
Deter m in a tion of Ster eoch em istr y. The stereo-
chemistry of the tandem adducts was determined by
either X-ray crystallography, conversion of the adducts
to known compounds, or comparison of HPLC and NMR
patterns. A single crystal of the major isomer of 1a , for
example, was obtained by recrystallization, and its X-ray
analysis unambiguously revealed the relative configura-
tion of 1a to be syn.18 Configurations of other compound
1 were determined by comparison of 1H NMR. For
example, H3 signals for syn-1 always appeared in the
lower field than the corresponding signals for anti-1.
Seleno analogues 2a , whose diastereomeric ratio was
81/19, were readily converted into known compound 9.
The 1H NMR spectrum of the minor isomer of 9 was
identical to the reported spectrum of anti-9 (Scheme 6).19
Thus, the configuration of major 2a was assigned to be
syn.
Sch em e 3
(Table 3, entry 6). In contrast, aliphatic aldehyde gave
only a trace amount of tandem adduct 4d (Table 3,
entry 5).
Th e Ta n d em Mich a el/Ald ol Rea ction to Cr oto-
n a te Ester s. Examination was made on the extension
of the reaction with crotonate esters (Scheme 5). The
results are summarized in Table 4.
The existence of the â-methyl group spoiled the elec-
trophilic reactivity of the crotonates. As a result, -50 °C
was too low for the reaction (Table 4, entry 1). To enhance
the reaction rate, the reaction mixture was allowed to
warm to room temperature and the tandem adduct 6a
was obtained in 55% yield (Table 4, entry 2). The
presence of three contiguous stereogenic centers in ad-
duct 6a gives rise to a possibility of the reaction forming
a mixture of four diastereomers (A-D), two of which were
obtained as major isomers (A and C) in a ratio of 53/1/
45/1 (Table 4, entry 2). Selenolate also gave analogue 7
in 62% yield at a similar diastereomeric ratio (Table 4,
entry 5). A better stereoselectivity was observed in the
reaction of 1-naphthyl aldehyde (Table 4, entry 3). An
aliphatic aldehyde was less reactive, forming only trace
amounts of the adduct (Table 4, entry 4).
Despite our effort, the adducts from methacrylate
never afforded any single crystal suitable for X-ray
analysis. An attempt was made to convert it to acetal
derivative 11 (Scheme 7).
Reduction of 4b with LiAlH4 gave diol 10, which was
then converted into acetal 11 by treatment with 2,2-
dimethoxypropane. An NOE experiment on 11 suggested
that the methyl group should be located in the cis position
to the two axial protons (Scheme 7) because 4.3% of signal
enhancement of the C2-methyl group was observed when
H3 nucleus was irradiated. X-ray analysis of a single
crystal of 11 proved this assignment to be true.20
The presence of an additional stereocenter in the
adducts made determination of stereochemistry of 6 and
7 less simple. As we mentioned above, X-ray crystal-
lographic analysis revealed the major isomer of 7 (7-A)
to have a 2S*,3R*,2RS* configuration. The second and
third major isomers (7-C and 7-D) were, however,
extremely difficult to isolate in diastereomerically pure
(17) Crystallographic data for 7-A have been deposited with the
Cambridge Crystallographic Data Centre.
Determination of the configurations for these adducts
was not easy. Fortunately, the major isomer (7-A) was
isolated as a single crystal, and its X-ray analysis clearly
revealed the stereochemistry of 7-A to be 2S*,3R*,2RS*.17
(18) Crystallographic data for syn-1a have been deposited with the
Cambridge Crystallographic Data Centre (deposit no. 114923).
(19) Corey, E. J .; Kim, S. S. J . Am. Chem. Soc. 1990, 112, 4976.
(20) Crystallographic data for 11 have been deposited with the
Cambridge Crystallographic Data Centre (deposit no. 114924).

6356 J . Org. Chem., Vol. 64, No. 17, 1999
Ta ble 2. Mich a el/Ald ol Ta n d em Ad d ition to Acr yla tes w ith Selen ola tes or Alk oxid es
Kamimura et al.
syn/antib
entry
R1X
R
solvent
T (°C)
-50
time (h)
product
yield (%)a
1
2
3
4
5
6
7
8
9
PhSe
Ph
ether
ether
7
15
15
7
2a
2b
2c
2c
2d
2e
2f
97
78
64
65
64
69
46
0
87/13
78/22
80/20
84/16
84/16
57/43
58/42
PhSe
PhSe
PhSe
PhSe
PhSe
PhSe
PhO
p-ClC6H4
1-Napc
1-Napc
2-Napc
PhCHdCH-
C5H11
-78 to rt
-78 to rt
-50
ether
CH2Cl2
ether
ether
ether
-78 to rt
-50
15
7
7
15
15
-50
-78 to rt
-78 to rt
Ph
Ph
CH2Cl2
CH2Cl2
3a
3b
PhCH2O
0
a
b
Isolated yield. Determined by HPLC analyses. c 1-Nap ) 1-naphthyl; 2-Nap ) 2-naphthyl.
Sch em e 4
Sch em e 6
Ta ble 3. Mich a el/Ald ol Ta n d em Ad d ition to ter t-Bu tyl
Meth a cr yla tes
T
(°C)
time
yield syn/
entry
R
X
solvent
(h) product (%)a antib
1
2
Ph
Ph
S
CH2Cl2 -15
7
4a
4a
61
67
87/13
83/17
S
S
S
S
CH2Cl2 -78 to 15
rt
3
4
5
6
p-ClC6H4
1-Napc
C5H11
Ph
CH2Cl2 -78 to 15
4b
4c
4d
5
55
89/11
80/20
-
Sch em e 7
rt
CH2Cl2 -78 to 15
39
rt
CH2Cl2 -78 to 15
trace
69
rt
Se ether
-78 to 15
rt
75/25
a
b
Isolated yield. Determined by HPLC analyses. c 1-Nap )
1-naphthyl.
Sch em e 5
the number of diastereomers of 7 to two. We had in hand
two kinds of mixtures of the four diastereoisomers of 7,
A/B/C/D ) 56/1/42/1 for one and 11/3/48/38 for the
other.21 Treatment of the former mixture (A/B/C/D ) 56/
1/42/1) with Bu3SnH gave a diastereomeric mixture of
12, whose ratio was 55/45. The same transformation of
the latter mixture (A/B/C/D ) 11/3/48/38) afforded 12 in
a ratio of 10/90 (Scheme 8). These results indicate that
diastereomers A and B have identical configurations
between C2 and C2R and so do C and D. Comparing the
X-ray result of 7-A, we concluded that diastereomers A
and B are syn-aldols and diastereomers C and D anti-
aldols.
We then tried to remove the stereogenic center of the
hydroxyl carbon C2R of 7. All the experiments to convert
7 into the corresponding ketone, however, failed due to
the lability of the phenylseleno group under oxidative
conditions. That each of the diastereoisomers of 6a and
7 shared almost the same NMR patterns suggested the
stereochemistry of 6a -A and 7-A to be identical. Instead,
we used thio derivative 6a for further transformation
(Scheme 9). The results are summarized in Table 5.
Ta ble 4. Mich a el/Ald ol Ta n d em Ad d ition to ter t-bu tyl
Cr oton a tes
T
(°C)
time
yield
entry
R
X
solvent
(h) product (%)a A/B/C/Db
1
2
Ph
Ph
S
CH2Cl2 -50
7
6a
6a
trace
55
S
S
S
CH2Cl2 -78 to 15
53/1/45/1
75/1/15/9
rt
3
4
5
1-Napc
C5H11
Ph
CH2Cl2 -78 to 15
6b
6c
7
60
rt
CH2Cl2 -78 to 15
trace
62
rt
Se ether
-78 to 15
rt
56/1/42/1
a
b
Isolated yield. Determined by HPLC analyses. c 1-Nap )
1-naphthyl.
form. Only the HPLC method separated them in small
amounts, not enough for this study to perform further
transformation. Instead, we used a mixture of diastere-
omers of 6 or 7 for the following study.
First of all, we attempted to remove the phenylseleno
group at C3 position in 7. This transformation reduced
(21) This mixture was prepared through the reaction with magne-
sium selenophenolate or thiophenolate which showed anti-aldol selec-
tivity under the same reaction conditions. Details on this matter will
be reported elsewhere.

Thio-Michael/Aldol Tandem Reactions
J . Org. Chem., Vol. 64, No. 17, 1999 6357
Sch em e 8
Sch em e 10
Sch em e 9
acrylate to the mixture seemingly solved a very small
part of the precipitate but left most of PhSLi where it
was. This was supported by the NMR experiments in
which no significant peaks except those came from
acrylate were detected in the NMR spectrum of the
heterogeneous mixture of the acrylate and thiolate in
CD2Cl2. Thus, active enolate intermediate B in the
solution phase is very low of concentration. On the other
hand, addition of aldehyde to the initial mixture of PhSLi
gave a similar result; a small part of the precipitate
seemed to be solved, but the mixture itself stayed as it
was. No significant change in the chemical shift at the
formyl proton was observed in the NMR spectrum of the
mixture of PhSLi-aldehyde in CD2Cl2.
A dramatic change, however, occurred when both
aldehyde and acrylate were added to the mixture: the
whole precipitate rapidly solved (less than 5 min at -50
°C), and the reaction mixture turned into a homogeneous
pale yellow solution. From this observation it is gathered
that the two carbonyl groups cooperatively coordinate to
the lithium cation so that they may form acrylate-
thiolate-aldehyde complex D soluble in CH2Cl2. Complex
D may be formed through a nucleophilic attack at
acrylate by the sulfur atom in complex C, or through
coordination of the aldehyde to complex B, but it is not
clear that which pathway is favored.23 Subsequent struc-
tural changes convert complex D to complex E, which
affords the aldol adduct syn-selectively.
Ta ble 5. Oxid a tion of 6a
ratio of 6a 13; yielda
14; yielda
entry (A/B/C/D)b
(%)
A/B/C/Db
(%)
14a /14bb
1
2
3
53/1/45/1
5/1/32/62
100/0/0/0
96
68
100
50/1/48/1
3/1/32/64
100/0/0/0
75
100
94
96/4
34/66
100/0
a
b
Isolated yield. Determined by HPLC analyses.
The oxidation started with two kinds of mixtures of
four diastereoisomers of 6a , their respective ratios A/B/
C/D ) 53/1/45/1 and 5/1/32/62.21 The phenylthio group
was first oxidized to sulfone 13, and the subsequent PCC
oxidation resulted in ketone 14 in good yields. The
diastereomeric ratios of the mixtures were kept up within
the experimental error during the oxidation (Table 5,
entries 1 and 2). Diastereomerically pure 6a -A was
converted into pure 14a in a quantitative yield (Table 5,
entry 3).22 The configuration of 14a was thus determined
to be 2S*,3R*. The former mixture afforded 14a and 14b
at the ratio of 96/4, and the latter did 14 at the ratio of
34/66. These results suggest that 6a -C has the same
configuration at C2 and C3 as does 6a -A, and that
opposite to it is the configuration of 6a -B and 6a -D.
Consequently, the configurations of all the four diaster-
eomers of 6a were determined to be 2S*,3R*,2RS* for
A, 2S*,3S*,2RS* for B, 2S*,3R*,2RR* for C, and
2S*,3S*,2RR* for D.
Ster eoch em ica l Cou r se of th e Rea ction . The tan-
dem addition to acrylates and methacrylates provides
syn-aldol adducts selectively, and the reaction mecha-
nism is assumed to be the way as it follows (Scheme 10).
Thiophenol in CH2Cl2 is deprotonated with butyl-
lithium. It gives lithium thiophenolate in a white pre-
cipitate, which should exist in polymeric form such as
(PhSLi)n A. The concentration of the thiophenolate anion
in the solution phase is very low. In the NMR spectrum
of the heterogeneous mixture of PhSLi in CD2Cl2, almost
no signals were observed in aromatic region. Addition of
When the same reaction is carried out in THF, a more
coordinative solvent, formation of complex D is impos-
sible because a large amount of THF molecules predomi-
nantly coordinate to the lithium cation, which loses its
Lewis acidity, and in fact, the yield and the selectivity
became poor (see Table 1, entry 10). Consequently, we
can conclude that the Lewis acidity of naked lithium
cation is the key driving force to let the reaction proceed
successfully.24 For this reason, it requires lithium cation
(23) Reversed order of the addition, i.e., the aldehyde first then the
acrylate, sometimes gave better yields of the tandem adducts, but the
stereoselectivity was almost identical.
(24) Grieco, P. A.; Kaufman, M. D.; Daeuble, J . F.; Saito, N. J . Am.
Chem. Soc. 1996, 118, 2095. Grieco, P. A.; Beck, J . P.; Handy, S. T.;
Saito, N.; Daeuble, J . F. Tetrahedron Lett. 1994, 35, 6783. Grieco, P.
A.; Nunes, J . J .; Gaul, M. D. J . Am. Chem. Soc. 1990, 112, 4595 and
references therein.
(22) No epimerization were observed during the transformation to
14; no trace amount of 14b was observed in the 1H NMR spectrum of
the crude reaction mixture starting from diastereomerically pure 6a -A
(Table 6, entry 3).

6358 J . Org. Chem., Vol. 64, No. 17, 1999
Kamimura et al.
Sch em e 11
Sch em e 12
pound 15 was then put under the standard radical
cyclization conditions (Bu3SnH/AIBN at 110 °C) and
yielded tetrahydrofuran 16 as the sole isomer. The
cyclization took place with a high stereoselectivity as had
been reported in the previous literature.29-31 The stereo-
chemistry of 16 was determined by the NOE experiment
in which the 2,3-trans-3,5-trans configuration was elu-
cidated because H5 and H3 signals were enhanced by
9.0% and 0.0%, respectively, when the H2 nucleus was
irradiated. This application proves the utility of the
tandem reaction in organic synthesis.
as the countercation and CH2Cl2 as the solvent for the
present tandem reaction to be successfully performed.
In regard to the reaction with crotonates, on the other
hand, the syn-aldol selectivity disappeared and the anti-
Michael selectivity dominated instead. Although the
reason this switching of selectivity is not yet made clear,
this anti-Michael selectivity is plausibly explained as
Scheme 11 shows. The first stage of the reaction is the
formation of three-component complex F , which is
achieved in a similar way that forms complex D in
Scheme 10. Complex F further changes into two confor-
mational isomers G and H. As with conformer G, due to
the steric bias brought by the phenylthio group, the
aldehyde preferentially attacks the top face of the enolate
giving an anti-Michael adduct. As with conformer H, the
aldehyde comes from the bottom face to give a syn-
Michael adduct. Conformer G should be a more favorable
conformation than conformer H in that the latter causes
a steric repulsion to arise between the methyl group and
the oxygen atom in the enolate unit. As a result, the aldol
reaction proceeds through the attack of the aldehyde at
the top face of conformer G, giving an anti-Michael
product selectively. Similar observations have been re-
ported in the Michael addition of thiolate anion to
acrylate esters and nitroolefins.25,26
Con clu sion
We have presented a new method of the Michael/aldol
tandem reaction among lithium thiolate (or selenolate),
acrylate, and aldehyde, all of which were condensed in
one pot, giving good yields of â-hydroxy-R-phenylthio- or
â-hydroxy-R-phenylselenoalkyl esters with a high ste-
reoselectivity. The reaction progress and selectivity
largely depend on the choice of solvents, countercations,
and the ester parts of acrylate. Since the present reaction
procedure readily creates multifunctional molecules be-
sides a new carbon-carbon bond, this will open up a new
method in organic synthesis. Further application and
mechanistic investigation on the reaction is now under-
way in our laboratory.
Exp er im en ta l Section
Gen er a l. All 1H and 13C NMR spectra were measured in
CDCl3 and recorded on J EOL EX-270 (270 MHz for 1H and
67.5 MHz for 13C) spectrometer. All the reactions in this paper
were performed under nitrogen atmosphere. Solvents used in
the reaction described here were dried over appropriate drying
agents (K for THF, Na for ether and toluene, and CaH2 for all
other solvents) and distilled under nitrogen before use. Alde-
hydes were purified by distillation. Methyl and ethyl acrylate,
Ster eoselective P r ep a r a tion of Tetr a h yd r ofu r a n .
To see the applicability of the present method, stereose-
lective preparation of trisubstituted tetrahydrofuran was
examined (Scheme 12).27
Seleno-tandem adduct 2a was O-alkenylated with
methyl propiolate and gave 15 in a good yield.28 Com-
(28) Winterfeldt, E.; Preuss, H. Chem. Ber. 1966, 99, 450.
(29) Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry of Radical
Reactions; VCH: Weinheim, 1996.
(25) Miyata, O.; Shinada, T.; Ninomiya, I.; Naito, T.; Date, T.;
Okamura, K.; Inagaki, S. J . Org. Chem. 1991, 56, 6556.
(26) Kamimura, A.; Ono, N. J . Chem. Soc., Chem. Commun. 1988,
1278. Kamimura, A.; Ono, N. Tetrahedron Lett. 1989, 30, 731.
Kamimura, A.; Sasatani, H.; Hashimoto, T.; Kawai, T.; Hori, K.; Ono,
N. J . Org. Chem. 1990, 55, 2437. Hori, K.; Higuchi, S.; Kamimura, A.
J . Org. Chem. 1990, 55, 5900.
(27) Reviews for tetrahydrofuran synthesis: Harmange, J .-C.; Fi-
gade`re, B. Tetrahedron: Asymmetry 1993, 4, 1711. Postema, M. H. D.
Tetrahedron 1992, 48, 8545. Boivin, T. L. B. Tetrahedron 1987, 43,
3309.
(30) Frauenrath, H.; Runsink, J . J . Org. Chem. 1987, 52, 2707.
Lindermann, R. J .; Godfrey, A. J . Am. Chem. Soc. 1988, 110, 6249.
Hoppe, D.; Kra¨mer, T.; Erdbru¨gger, C. F.; Egert, E. Tetrahedron Lett.
1989, 30, 1233. Hopkins, M. H.; Overman, L. E.; Rishton, G. M. J .
Am. Chem. Soc. 1991, 113, 5354; Brown, M. J .; Harrison, T.; Herrinton,
P. M.; Hopkins, M. H.; Hutchinson, K. D.; Mishra, P.; Overman, L. E.
Ibid. 1991, 113, 5365. Brown, M. J .; Harrison, T.; Overman, L. E. Ibid.
1991, 113, 5378.
(31) Lee, E.; Park, C. M. J . Chem. Soc., Chem. Commun. 1994, 293.

Thio-Michael/Aldol Tandem Reactions
J . Org. Chem., Vol. 64, No. 17, 1999 6359
which were purchased from Nakalai Tesque, tert-butyl acry-
late, Bu3SnH, thiophenol, and diphenyldiselenide, which were
from Aldrich, were used without further purification. Other
R,â-unsaturated esters were prepared from corresponding acyl
chloride and alcohols or acid and isobutene.
cm-1. Anal. Calcd for C21H26O3S: C, 70.36; H, 7.31. Found:
C, 70.13; H, 7.53.
Mich a el/Ald ol Ta n d em Rea ction to r,â-Un sa tu r a ted
Ester s w ith Selen ola te An ion . P r ep a r a tion of ter t-Bu tyl
3-H y d r o x y -3-p h e n y l-2-(p h e n y ls e le n o m e t h y l)-p r o p i-
on a te (syn -2a ). Gen er a l p r oced u r e. Methyllithium in ether
(1.1 M, 3.0 mL, 3.3 mmol) was added to a solution of
diphenyldiselenide (1.03 g, 3.30 mmol) in ether (6 mL) at room
temperature until the yellow color of the diselenide disap-
peared. The colorless solution was maintained at the same
temperature for 30 min and then cooled to -50 °C. Benzal-
dehyde (0.350 g, 3.30 mmol) was added to the mixture, and
the resulting mixture was allowed to stir for 10 min. The
reaction mixture turned to homogeneous pale yellow solution.
To the solution was added tert-butyl acrylate (0.385 g, 3.00
mmol), and the reaction mixture was allowed to stir at -50
°C for 7 h. Aqueous HCl (1 M, 10 mL) was added, and the
resulting mixture was extracted with ethyl acetate (3 × 30
mL). The organic phases were combined and dried over Na2-
SO4. After filtration and removal of the solvent in vacuo, crude
product was purified with flash chromatography (silica gel/
hexanes-ether 20:1 then 3:1 v/v) to give desired tandem
product 2a in 97% yield (1.126 g, 2.88 mmol) as a pale yellow
oil. HPLC analysis indicated disatereomeric ratio was 87/13:
1H NMR (270 MHz, CDCl3) δ 7.19-7.36 (m, 10 H), 4.98 (d, 1
H, J ) 5.6 Hz), 3.14 (d, 1 H, J ) 6.3 Hz), 3.14 (d, 1 H, J ) 7.6
Hz), 2.96 (dt, 1 H, J ) 5.6, 7.6 Hz), 2.8-3.0 (br, 1 H), 1.35 (s,
9 H); 13C NMR (67.5 MHz, CDCl3) δ 172.6, 140.8, 132.3, 129.0,
128.4, 128.3, 127.9, 126.8, 126.3, 81.9, 74.6, 54.0, 27.9, 24.4;
IR 3620-3100, 1725 cm-1. Anal. Calcd for C20H24O3Se: C,
61.38; H, 6.18. Found: C, 61.26; H, 6.13.
Mich a el/Ald ol Ta n d em Rea ction to r,â-Un sa tu r a ted
Ester s w ith Th iola te An ion . P r ep a r a tion of ter t-Bu tyl
3-Hydr oxy-3-ph en yl-2-(ph en ylth iom eth yl)pr opion ate (syn -
1d ). Gen er a l P r oced u r e. To a solution of thiophenol (0.242
g, 2.20 mmol) in CH2Cl2 (2 mL) was added butyllithium in
hexanes (1.1 M, 2 mL, 2.2 mmol) at -78 °C, and lithium
thiophenolate precipitated as white solid. To the heterogeneous
mixture were added tert-butyl acrylate (0.256 g, 2.00 mmol)
and benzaldehyde (0.212 g, 2.00 mmol) at -78 °C; the reaction
mixture turned to a pale yellow solution, which was main-
tained at -50 °C for 7 h. Aqueous HCl (1M, 5 mL) was added,
and the mixture was extracted with ethyl acetate (3 × 30 mL).
The organic phase was washed with brine (10 mL) and dried
over Na2SO4. After filtration and removal of the solvent in
vacuo, crude product was purified with flash chromatography
(silica gel/hexanes-ether 20:1 then 3:1 v/v) and the desired
tandem product 1d was obtained in 80% yield (0.548 g, 1.59
mmol) as pale yellow oil. HPLC analysis indicated disatereo-
meric ratio was 92/8: 1H NMR (270 MHz, CDCl3) δ 7.14-7.33
(m, 10 H), 4.97 (d, 1 H, J ) 5.6 Hz), 3.21 (d, 1 H, J ) 5.6 Hz),
3.20 (d, 1 H, J ) 8.3 Hz), 2.86 (td, 1 H, J ) 5.6, 7.6 Hz), 2.83-
2.93 (br, 1 H), 1.33 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ
172.4, 140.8, 135.8, 129.2, 128.8, 128.3, 127.9, 126.3, 126.0,
81.8, 74.2, 53.2, 31.2, 27.9; IR 3100-3700, 1720 cm-1. Anal.
Calcd for C20H24O3S: C, 69.74; H, 7.02. Found: C, 69.35; H,
7.08.
The two diastereomers of 1a and 1b were separated by
careful flash column chromatography (hexanes-ether 5:1 v/v).
Meth yl 3-h yd r oxy-3-p h en yl-2-(p h en ylth iom eth yl)p r o-
p ion a te (syn -1a ): mp 59 °C (from hexane-CH2Cl2); 1H NMR
(270 MHz, CDCl3) δ 7.11-7.37 (m, 10 H), 5.04 (dd, 1 H, J )
2.6, 5.0 Hz), 3.57 (s, 3 H), 3.26 (dd, 1 H, J ) 9.2, 13.9 Hz),
3.20 (dd, 1 H, J ) 4.6, 13.9 Hz), 2.97 (td, 1 H, J ) 5.0, 9.3 Hz),
2.78 (d, 1 H, J ) 3.0 Hz); 13C NMR (67.5 MHz, CDCl3) δ
173.4, 140.7, 135.3, 129.3, 128.9, 128.4, 128.0, 126.2, 125.9,
74.0, 53.0, 51.9, 30.7; IR 3200-3700, 1730 cm-1. Anal. Calcd
for C17H18O3S: C, 67.52; H, 6.00. Found: C, 67.24; H, 6.01.
Meth yl 3-h yd r oxy-3-p h en yl-2-(p h en ylth iom eth yl)p r o-
ter t-Bu tyl 3-h yd r oxy-2-m eth yl-3-p h en yl-2-(p h en ylse-
len om eth yl)p r op ion a te (syn -5): colorless oil; 1H NMR (270
MHz, CDCl3) δ 7.20-7.48 (m, 10 H), 4.98 (s, 1 H), 3.40 (d, 1
H, J ) 12.2 Hz), 2.93 (d, 1 H, J ) 12.5 Hz), 1.45 (s, 9 H), 1.18
(s, 3 H). 13C NMR (67.5 MHz, CDCl3) δ 174.8, 139.7, 132.5,
129.3, 129.2, 128.2, 128.1, 127.9, 126.9, 82.5, 78.8, 53.2, 33.7,
28.2, 20.6; IR 3100-3650, 1720 cm-1. Anal. Calcd for C21H26O3-
Se: C, 62.22; H, 6.46. Found: C, 62.35; H, 6.51.
ter t-Bu tyl (2S*,3R*)-2-((S*)-h yd r oxyp h en ylm eth yl)-3-
p h en ylselen obu ta n oa te (7-A): mp 111 °C; 1H NMR (270
MHz, CDCl3) δ 7.27-7.64 (m, 10 H), 5.21 (dd, 1 H, J ) 4.0,
9.2 Hz), 3.75 (dq, 1 H, J ) 4.0, 7.3 Hz), 2.89 (dd, 1 H, J ) 4.0,
9.2 Hz), 2.43 (1 H, d, J ) 4.0 Hz), 1.58 (d, 3 H, J ) 7.3 Hz),
1.23 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ 170.4, 141.8, 134.1,
129.0, 128.4, 128.2, 128.1, 127.3, 127.1, 81.6, 74.4, 59.9, 39.9,
27.8, 22.5; IR 3300-3600, 1720 cm-1. Anal. Calcd for C21H26O3-
Se: C, 62.22; H, 6.46. Found: C, 62.05; H, 6.39.
1
p ion a te (a n ti-1a ): colorless oil; H NMR (270 MHz, CDCl3)
δ 7.18-7.37 (m, 10 H), 4.96 (t, 1 H, J ) 5.6 Hz), 3.63 (s, 3H),
3.05 (dd, 1 H, J ) 3.6, 10.2 Hz), 3.00 (1H, dd, J ) 6.6, 10.2
Hz), 2.94-3.03 (m, 1 H), 2.95 (d, 1 H, J ) 5.6 Hz); 13C NMR
(67.5 MHz, CDCl3) δ 173.8, 141.0, 134.9, 130.1, 128.9, 128.6,
128.2, 126.6, 126.0, 74.2, 52.6, 52.0, 33.2; IR 3200-3700, 1730
cm-1. Anal. Calcd for C17H18O3S: C, 67.52; H, 6.00. Found:
C, 67.22; H, 5.98.
ter t-Bu tyl (2S*,3R*)-2-((R*)-h yd r oxyp h en ylm eth yl)-3-
p h en ylselen o-bu ta n oa te (7-C): pale yellow oil; 1H NMR (270
MHz, CDCl3) δ 7.23-7.64 (m, 10 H), 5.06 (d, 1 H, J ) 5.6 Hz),
3.6-3.8 (br, 1 H), 3.30 (dq, 1 H, J ) 6.9, 7.3 Hz), 2.80 (dd, 1
ter t-Bu tyl
3-h yd r oxy-2-m eth yl-3-p h en yl-2-(p h en yl-
th iom eth yl)p r op ion a te (syn -4a ): colorless oil; 1H NMR (270
MHz, CDCl3) δ 7.14-7.38 (m, 10 H), 4.96 (s, 1 H), 3.45 (d, 1
H, J ) 12.2 Hz), 3.27 (br, 1 H), 3.08 (d, 1 H, J ) 12.5 Hz), 1.43
(s, 9 H), 1.18 (s, 3 H); 13C NMR (67.5 MHz, CDCl3) δ 174.1,
139.6, 137.9, 129.2, 128.8, 127.9, 127.8, 127.6, 125.9, 82.1, 52.9,
41.6, 39.8, 27.9, 18.6; IR 3100-3700, 1720 cm-1. Anal. Calcd
for C21H26O3S: C, 70.36; H, 7.31. Found: C, 70.94; H, 7.04.
ter t-Bu tyl (2S*,3R*)-2-((S*)-h yd r oxyp h en ylm eth yl)-3-
p h en ylth iobu ta n oa te (6a -A): mp 101-102 °C; 1H NMR (270
MHz, CDCl3) δ 7.22-7.60 (m, 10 H), 5.20 (dd, 1 H, J ) 3.6,
8.9 Hz), 3.79 (dq, 1 H, J ) 4.6, 7.3 Hz), 2.95 (dd, 1 H, J ) 4.3,
8.9 Hz), 2.63 (d, 1 H, J ) 3.6 Hz), 1.43 (d, 3 H, J ) 7.3
Hz), 1.23 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ 170.2, 141.8,
136.0, 131.6, 128.9, 128.3, 128.1, 127.1, 126.8, 81.4, 73.4, 59.0,
43.6, 27.7, 21.0; IR 3300-3600, 1720 cm-1. Anal. Calcd for
H, J ) 5.9, 7.9 Hz), 1.55 (d, 3 H, J ) 6.9 Hz), 1.35 (s, 9 H); 13
C
NMR (67.5 MHz, CDCl3) δ 172.7, 141.7, 135.1, 128.9, 128.7,
128.2, 127.6, 127.5, 126.0, 82.3, 73.0, 58.9, 38.4, 27.9, 21.5; IR
3300-3600, 1720 cm-1. Anal. Calcd for C21H26O3Se: C, 62.22;
H, 6.46. Found: C, 62.23; H, 6.67.
ter t-Bu tyl (2S*,3S*)-2-((S*)-h yd r oxyp h en ylm eth yl)-3-
p h en ylselen o-bu ta n oa te (7-D):. pale yellow oil; 1H NMR
(270 MHz, CDCl3) δ 7.47-7.51 (m, 2 H), 7.15-7.27 (m, 8 H),
5.24 (d, 1 H, J ) 4.3 Hz), 3.6-3.7 (br, 1 H), 3.45 (dq, 1 H, J )
6.9, 7.3 Hz), 2.72 (dd, 1 H, J ) 5.0, 8.6 Hz), 1.40 (d, 3 H, J )
6.9 Hz), 1.19 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ 172.8,
141.9, 135.5, 129.0, 128.4, 128.2, 127.9, 127.4, 125.7, 82.2, 73.1,
58.2, 37.9, 27.8, 20.4; IR 3300-3600, 1720 cm-1
.
Red u ctive Rem ova l of P h en ylselen o Gr ou p fr om Ta n -
d em Ad d u ct 2a . P r ep a r a tion of ter t-Bu tyl 3-Hyd r oxy-2-
m eth yl-3-p h en ylp r op ion a te 9. A solution of 2a (0.232 g, 0.5
mmol, 81/19 mixture of syn-2a and anti-2a ), Bu3SnH (0.291
g, 1.00 mmol), and AIBN (0.016 g, 0.10 mmol) in toluene (2
mL) was heated at 110 °C for 2.5 h. The resulting mixture
was subjected to flash chromatography (silica gel/hexane ether
10:1 then 3:1 v/v) to give 912,19 as a colorless oil in 75% yield
(0.116 g, 0.380 mmol). RP HPLC analysis showed diastereo-
C
21H26O3S: C, 70.36; H, 7.31. Found: C, 70.37; H, 7.51.
ter t-Bu tyl (2S*,3R*)-2-((R*)-h yd r oxyp h en ylm eth yl)-3-
p h en ylth iobu ta n oa te (6a -C): colorless oil; 1H NMR (270
MHz, CDCl3) δ 7.20-7.60 (m, 10 H), 5.03 (d, 1 H, J ) 5.0 Hz),
3.43 (qd, 1 H, J ) 6.9, 8.9 Hz), 2.76 (dd, 1 H, J ) 5.0, 8.9 Hz),
1.43 (d, 3 H, J ) 7.3 Hz), 1.30 (s, 9 H); 13C NMR (67.5 MHz,
CDCl3) δ 172.8, 141.8, 134.0, 133.0, 128.8, 128.2, 127.5, 126.8,
125.8, 82.2, 72.2, 58.0, 43.5, 27.8, 20.0; IR 3300-3600, 1720

6360 J . Org. Chem., Vol. 64, No. 17, 1999
Kamimura et al.
meric ratio of 9 was 81/19. Anal. Calcd for C14H20O3: C, 71.16;
H, 8.53. Found: C, 70.71; H, 8.61.
169.8, 141.1, 137.5, 133.9, 129.1, 128.9, 128.2, 128.0, 127.1,
81.8, 72.2, 60.4, 52.7, 27.4, 12.2; IR 3150-3700, 1730 cm-1
.
For syn-9 (major isomer): 1H NMR (270 MHz, CDCl3) δ
7.26-7.35 (m, 5 H), 5.03 (d, 1 H, J ) 4.3 Hz), 3.13 (br, 1 H),
2.69 (dq, 1 H, J ) 4.3, 7.3 Hz), 1.40 (s, 9 H), 1.11 (d, 3 H, J )
7.3 Hz); 13C NMR (67.5 MHz, CDCl3) δ 175.3, 141.5, 128.1,
127.4, 126.1, 81.1, 73.8, 47.0, 27.9, 11.0.
Anal. Calcd for C21H26O5S: C, 64.59; H, 6.71. Found: C, 64.50;
H, 6.93.
ter t-Bu tyl (2S*,3R*)-3-Ben zen esu lfon yl-2-ben zoylbu -
ta n oa te 14a . To a solution of 13 (diastereomeric ratio ) 50:
1:48:1, 0.21 g, 0.54 mmol) in CH2Cl2 was added PCC (0.182 g,
0.80 mmol), and the resulting suspension was allowed to stir
for 4 h at ambient temperature. The supernatant of the
reaction mixture was subjected to column chromatography
(silica gel-hexane ether 10:1 then 1:1 v/v) to give 14 in 75%
yield (0.158 g, 0.410 mmol): white crystal; diastereomertic
ratio of 14 was 96:4; mp 76-77 °C; 1H NMR (270 MHz, CDCl3)
δ 8.06 (d, 2 H, J ) 7.3 Hz), 7.94 (d, 2 H, J ) 7.3 Hz), 7.48-
7.70 (m, 6 H), 4.93 (d, 1 H, J ) 8.9 Hz), 4.20 (td, 1 H, J ) 7.3,
8.9 Hz), 1.37 (s, 9 H), 1.22 (d, 3 H, J ) 7.3 Hz); 13C NMR (67.5
MHz, CDCl3) δ 192.9, 165.8, 137.2, 136.3, 133.9, 129.1, 129.0,
128.9, 128.7, 128.6, 82.9, 59.4, 52.7, 27.5, 12.3; IR 1740, 1730
cm-1. Anal. Calcd for C21H24O5S: C, 64.93; H, 6.23. Found:
C, 64.67; H, 6.34.
For anti-9 (minor isomer): 1H NMR (270 MHz, CDCl3) δ
7.26-7.35 (m, 5 H), 4.70 (d, 1 H, J ) 7.9 Hz), 3.22 (br, 1 H),
2.71 (quint, 1 H, J ) 7.3 Hz), 1.44 (s, 9 H), 1.01 (d, 3 H, J )
7.3 Hz); 13C NMR (67.5 MHz, CDCl3) δ 175.3, 141.8, 128.3,
127.8, 126.6, 81.2, 73.8, 47.7, 27.9, 14.9.
Conversion of 7 to 12 was carried out in the same way.
ter t-Bu tyl 2-(h yd r oxyp h en ylm eth yl)bu ta n on a te (syn -
12): colorless oil; 1H NMR (270 MHz, CDCl3) δ 7.16-8.06 (m,
5 H), 4.81 (d, 1 H, J ) 6.9 Hz), 2.47 (ddd, 1 H, J ) 1.3, 2.0, 7.3
Hz), 1.40-1.69 (m, 2 H), 1.47 (s, 9 H), 0.83 (dt, 1 H, J ) 2.0,
7.3 Hz); 13C NMR (67.5 MHz, CDCl3) δ 174.2, 141.8, 128.1,
127.5, 126.4, 80.9, 74.4, 55.2, 27.9, 23.6, 11.9. Anal. Calcd for
C15H22O3: C, 71.97; H, 8.86. Found: C, 71.86; H, 9.25.
ter t-Bu tyl 2-(h yd r oxyp h en ylm eth yl)bu ta n on a te (a n ti-
12): colorless oil; 1H NMR (270 MHz, CDCl3) δ 7.24-7.65 (m,
5 H), 4.76 (t, 1 H, J ) 6.6 Hz), 3.12 (d, 1 H, J ) 6.3 Hz), 2.58
(ddd, 1 H, J ) 5.3, 7.3, 9.3 Hz), 1.40-1.69 (m, 2 H), 1.40 (s, 9
H), 0.91 (t, 3 H, J ) 7.3 Hz); 13C NMR (67.5 MHz, CDCl3) δ
174.6, 142.3, 128.2, 127.5, 126.3, 81.0, 74.9, 55.1, 27.9, 22.8,
11.4. Anal. Calcd for C15H22O3: C, 71.97; H, 8.86. Found: C,
71.51; H, 9.18.
ter t-Bu t yl (2S*,3S*)-3-b en zen esu lfon yl-2-b en zoylb u -
1
ta n oa te (14b): mp 100-101 °C; H NMR (270 MHz, CDCl3)
δ 7.45-8.01 (m, 10 H), 4.90 (d, 1 H, J ) 8.3 Hz), 4.32 (td, 1 H,
J ) 7.9, 7.3 Hz), 1.44 (d, 3 H, J ) 7.3 Hz), 1.34 (s, 9H); 13C
NMR (67.5 MHz, CDCl3) δ 191.3, 165.4, 137.6, 135.7, 133.9,
133.7, 129.1, 129.0, 128.9, 128.6, 83.4, 58.4, 54.0, 27.6, 11.7;
IR 1740, 1680 cm-1
.
Con ju ga te Ad d ition of 2a to Meth yl P r op iola te. To a
solution of 2a (0.587 g, 1.50 mmol) in CH2Cl2 (10 mL) were
added Et3N (0.22 mL, 1.6 mmol) and methyl propiolate (0.504
g, 6.00 mmol), and the reaction mixture was allowed to stand
at room temperature for 12 h. After removal of solvent, the
residue was subjected to flash chromatography (silica gel/
hexanes-ether 20:1 v/v) to give 15 as colorless oil in 88% yield
(0.630 g): 1H NMR (270 MHz, CDCl3) δ 7.41 (1 H, d, J ) 12.2
Hz), 7.16-7.38 (m, 10 H), 5.22 (d, 1 H, J ) 12.5 Hz), 5.06 (d,
1 H, J ) 7.6 Hz), 3.63 (s, 3 H), 3.25 (d, 1 H, J ) 4.0, 12.2 Hz),
3.17 (dd, 1 H, J ) 9.9, 11.9 Hz), 3.06 (ddd, 1 H, J ) 4.0, 7.6,
9.6 Hz), 1.25 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ 169.8,
167.7, 160.7, 136.8, 132.7, 132.4, 129.0, 128.8, 128.6, 126.9,
126.9, 98.9, 84.7, 81.8, 54.2, 51.0, 27.7, 24.9; IR 2980, 1730,
1540 cm-1. Anal. Calcd for C24H28O5Se: C, 60.63; H, 5.94.
Found:C, 60.62; H, 5.96.
P r ep a r a tion of 2,3-tr a n s-3,5-tr a n s-3-(ter t-Bu toxyca r -
bon yl)-5-(m eth oxyca r bon ylm eth yl)-2-p h en yltetr a h yd r o-
fu r a n (16). A mixture of 14 (0.452 g, 0.950 mmol), Bu3SnH
(0.395 g, 1.20 mmol), and AIBN (0.035 g, 0.20 mmol) in toluene
(10 mL) was heated under refluxing conditions for 3 h. After
the mixture was cooled, solvent was removed and the residue
was subjected into flash chromatography (silica gel/hexanes-
ethyl acetate 10:1 v/v) to give 16 as colorless oil in 90% yield
(0.229 g, 0.900 mmol): 1H NMR (270 MHz, CDCl3) δ 7.28-
7.37 (m, 5 H), 5.00 (d, 1 H, 7.6 Hz), 4.54 (quint, 1 H, J ) 6.9
Hz), 3.72 (s, 3 H), 2.89 (ddd, 1 H, J ) 6.3, 7.6, 13.5 Hz), 2.80
(dd, 1 H, J ) 6.9, 15.2 Hz), 2.64 (dd, 1 H, J ) 6.3, 15.5 Hz),
2.49 (quint, 1 H, J ) 6.6 Hz), 2.00 (ddd, 1 H, J ) 7.6, 9.6, 12.9
Hz), 1.43 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ 172.1, 171.1,
140.9, 128.9, 127.6, 125.8, 83.7, 75.2, 53.0, 51.6, 40.1, 35.5, 27.9,
27.3; IR 2950, 1710 cm-1. Anal. Calcd for C18H24O5: C, 67.48;
H, 7.55. Found: C, 67.24; H, 7.79.
Red u ction of 4b. To a suspension of LiAlH4 (0.284 g, 7.48
mmol) in THF (10 mL) was added 4b (0.733 g, 1.87 mmol) at
room temperature over 30 min, and the resulting mixture was
heated at refluxing temperature for 1 h. Diluted HCl (1 M, 10
mL) was slowly added to the ice-cooled reaction mixture, and
THF in the biphasic mixture was removed with a rotary
evaporator. The water phase was then extracted with EtOAc
(3 × 30 mL), and the organic phase was combined and dried
over Na2SO4. After removal of Na2SO4 and solvent, crude
product 10 was purified by flash chromatography (silica gel/
hexanes-ethyl acetate 10:1 then 5:1 v/v): pale yellow oil; 60%
yield (0.323 g, 1.12 mmol); 1H NMR (270 MHz, CDCl3) δ 7.11-
7.31 (m, 5 H), 4.81 (s, 1 H), 3.73 (d, 1 H, J ) 11.5 Hz), 3.45 (d,
1 H, J ) 10.9 Hz), 3.32 (d, 1 H, J ) 12.2 Hz), 2.78 (d, 1 H, J
) 12.5 Hz), 1.50-1.60 (br, 2 H), 0.80 (s, 3 H); 13C NMR (67.5
MHz, CDCl3) δ 138.9, 133.5, 129.4, 129.0, 128.9, 128.1, 126.2,
77.5, 68.8, 43.5, 37.6, 18.5; IR 3300-3700 cm-1. Anal. Calcd
for C17H19ClO2S: C, 63.25; H, 5.93. Found: C, 63.11; H, 6.30.
Aceta liza tion of 10. A solution of 10 (0.131 g, 4.80 mmol),
2,2-dimethoxypropane (0.50 g, 4.8 mmol), and PPTS (0.025 g,
0.10 mmol) in toluene (2 mL) was heated under refluxing
conditions for 1 h. The reaction mixture was then concentrated
in vacuo, and the crude product was subjected to flash
chromatography (silica gel/hexanes-ethyl acetate 10:1 v/v) to
give 11 as a white solid in 100% yield (0.152 g, 4.80 mmol):
1
mp 96-97 °C; H NMR (270 MHz, CDCl3) δ 7.10-7.35 (m, 9
H), 4.81 (s, 1 H), 4.04 (d, 1 H, J ) 12.2 Hz), 3.66 (dd, 1 H, J )
1.7, 12.2 Hz), 3.58 (d, 1 H, J ) 12.5 Hz), 2.50 (dd, 1 H, J )
1.7, 12.9 Hz), 1.53 (s, 6 H), 0.82 (s, 3 H); 13C NMR (67.5 MHz,
CDCl3) δ 136.5, 129.7, 129.6, 129.5, 129.4, 128.7, 128.6, 126.4,
100.0, 80.2, 67.4, 38.5, 36.5, 30.4, 19.7, 19.4. Anal. Calcd for
C
20H23ClO2S: C, 66.19; H, 6.39. Found: C, 66.47; H, 6.69.
Oxid a tion of 6a . A solution of 6a (A:B:C:D ) 53:1:45:1,
Ack n ow led gm en t. We acknowledge financial sup-
port for this work from the UBE foundation (to A.K.),
Ube Industries Ltd., and Sumitomo Chemical Industry
Ltd. The present work was partially supported by a
Grant-in-Aid (11640536) from the Ministry of Educa-
tion, Science and Culture, J apan.
0.257 g, 0.710 mmol) and m-CPBA (80%, 0.270 g, 1.56 mmol)
in chloroform (10 mL) was heated to refluxing temperature
for 5 h. The reaction mixture was condensed under reduced
pressure, and the crude solution in chloroform was subjected
into flash chromatography (silica gel/hexanes-ethyl acetate
5:1 then 3:1 v/v) to give 13 in 96% yield (0.266 g, 0.680 mmol).
The diastereomertic ratio of 13 was 50:1:48:1. Oxidation
carried out with diastereomerically pure 6a -A gave a single
isomer of 13 as viscous oil.
Su p p or t in g In for m a t ion Ava ila b le: Physical data for
compounds 1b,c,f -l, 4b,c, 6b, and 2b-f, spectroscopic charts
for compounds 2f, 4a , and 14b, and ORTEP drawings for
structures syn-1a , 7-A, and 11. This material is available free
of charge via the Internet at http://pubs.acs.org.
ter t-Bu tyl (2S*,3R*)-2-((S*)-h yd r oxyp h en ylm eth yl)-3-
ben zen esu lfon ylbu tan oate (13a): 1H NMR (270 MHz, CDCl3)
δ 7.27-7.96 (m, 10 H), 5.04 (dd, 1 H, J ) 3.0, 8.6 Hz), 3.73
(td, 1 H, J ) 4.6, 7.3 Hz), 3.32 (dd, 1 H, J ) 4.6, 8.6 Hz), (d, 1
H, J ) 7.3 Hz), 1.18 (s, 9 H); 13C NMR (67.5 MHz, CDCl3) δ
J O990548S