A divergent synthesis of essentially enantiopure syn- and anti-propionate aldol adducts based on the chiral 1,3,2-oxazaborolidin-5-one-promoted asymmetric aldol reaction followed by diastereoselective radical reduction

Article abstract of DOI:10.1246/bcsj.74.1485

Essentially, enantiopure syn- and anti-propionate aldol adducts were divergently synthesized using a novel strategy which utilizes both the highly enantioselective 1,3,2-oxazaborolidin-5-one-promoted aldol reaction with a ketene silyl acetal derived from ethyl 2-bromo propionate and a highly diastereoselective radical debromination reaction. These procedures provide yields that increase to a level available for practical synthesis.

Full text of DOI:10.1246/bcsj.74.1485

c. Jpn.
© 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 1485–1495 (2001) 1485
e Chemical
ciety of Ja-
n
e Chemical
ciety of Ja-
n
A Divergent Synthesis of Essentially Enantiopure syn- and anti-Propi-
onate Aldol Adducts Based on the Chiral 1,3,2-Oxazaborolidin-5-one-
Promoted Asymmetric Aldol Reaction Followed by Diastereoselective
Radical Reduction
01
85
Divergent Synthesis of syn- and anti-Propionate
S. Kiyooka et al.
95
Syun-ichi Kiyooka* and Kazi Abuds Shahid
SJA8
09-2673
064
Department of Chemistry, Kochi University, Akebono-cho, Kochi 780-8520
(Received March 1, 2001)
01
Essentially, enantiopure syn- and anti-propionate aldol adducts were divergently synthesized using a novel strategy
which utilizes both the highly enantioselective 1,3,2-oxazaborolidin-5-one-promoted aldol reaction with a ketene silyl
acetal derived from ethyl 2-bromo propionate and a highly diastereoselective radical debromination reaction. These pro-
cedures provide yields that increase to a level available for practical synthesis.
01
0
The enantioselective synthesis of syn- and anti-propionate
aldol adducts is important in terms of the asymmetric synthesis
of natural products which contain 1,3-diol moieties. Consider-
able success has been reported in the area of asymmetric aldol
reactions with chiral nucleophiles which are related to propi-
onate synthons and which are bound to a variety of chiral aux-
iliaries.1 The chiral Lewis acid-promoted asymmetric aldol re-
action, in which activated aldehydes react with silyl nucleo-
philes, is generally considered to be more practical and useful
for the enantioselective construction of aldol adduct’s skele-
tons, compared to asymmetric aldol reactions which involve
chiral auxiliaries. In cases where a chiral Lewis acid can be
successfully used as a catalyst, asymmetric aldol reactions are
the reactions of choice. Even in the case where the chiral
Lewis acid does not function as a superior catalyst, the semi-
catalytic or stoichiometric process is still quite useful (in such
cases the Lewis acid is referred to as a promoter) and is more
advantageous situations that require the binding and removal
of chiral auxiliaries. A number of successful examples have
been reported in the area of the chiral Lewis acid-catalyzed
(promoted) enantioselective aldol reactions.2 However, the di-
astereoselectively divergent synthesis of essentially enan-
tiopure syn- and anti-propionate aldol adducts using chiral
Lewis acid-mediated asymmetric aldol reactions has not yet
been realized. In the case of chiral borane-catalyzed (promot-
ed) aldol reactions with ketene silyl acetals, as these relate to
propionate synthons, Yamamoto reported on a highly enanti-
oselective synthesis of syn-propionate aldol adducts with chiral
(acyloxy)borane reagents, derived from L- or D-tartaric acid, in
which syn predominance is obtained independently of the ge-
ometry of the ketene silyl acetals, while the reaction using
chiral 1,3,2-oxazaborolidin-5-ones resulted in the formation of
a slight excess of anti-propionate aldol adducts with moderate
enantioselectivity.3 The resolution of divergent synthesis rep-
resents the most important serious problem remaining since
our first report of the chiral 1,3,2-oxazaborolidin-5-one-medi-
Scheme 1. A plausible mechanism.
ated asymmetric aldol reaction.4
Our chiral 1,3,2-oxazaborolidin-5-one-mediated asymmet-
ric aldol reaction proceeds smoothly through a catalytic cycle,
as shown in Scheme 1, which represents the most plausible
mechanism. The supposed aldolate intermediate also appears
to catalyze the aldol reaction because it is a trivalent neutral
borane species, but in a somewhat lower enantioselectivity, as
observed when a catalytic amount of 1,3,2-oxazaborolidin-5-
one was used. In order to prevent a lowering of the enantiose-
lectivity, 1,3,2-oxazaborolidin-5-one, in which the chiral
ligands are recycled in nearly all cases, was used in a stoichio-
metric amount. The si facial selection is always observed in
the reaction using (S)-1,3,2-oxazaborolidin-5-one.4
We report herein a new approach toward a diastereoselec-
tively divergent synthesis of essentially enantiopure syn- and
anti-propionate aldol adducts using the chiral 1,3,2-oxazaboro-
lidin-5-one-promoted asymmetric aldol reaction, in conjunc-
tion with diastereoselective radical reduction.5

1486 Bull. Chem. Soc. Jpn., 74, No. 8 (2001)
Divergent Synthesis of syn- and anti-Propionate
Scheme 2. The combination of enantioselective aldol reaction and diastereoselective radical reduction toward enantiopure syn- and
anti-propionate aldol adducts.
tiopure aldol adducts.6 Fortunately, the radical-mediated hy-
Results and Discussion
drogen-transfer reactions, reported by Guindon, represent a
The appropriate steric bulkiness at the β-position of ketene
silyl acetals is known to allow a high level of enantioselectivity
in our chiral 1,3,2-oxazaborolidin-5-one-promoted aldol reac-
tion; for example, a reaction with a ketene silyl acetal bearing a
dithiolane resulted in the formation of essentially enantiopure
acetate aldol adducts after desulfurization.4g Based on this
strategy, a ketene silyl acetal was prepared from ethyl 2-(meth-
ylthio)propionate, which gave a mixture of an equal amount of
syn- and anti-aldol products having an α-methylthio group
with very high level of enantioselectivity in the aldol reaction,
followed by desulfurization to afford the essentially enan-
tiopure syn- and anti-propionate aldol adducts, but with poor
diastereoselection.4l At that stage, the earlier studies were
strongly in need of an effective diastereoselective desulfuriza-
tion procedure, or its equivalent. An ideal approach to realiz-
ing both enantiopure syn- and anti-propionate aldol adducts is
as follows. The asymmetric aldol reaction results in the forma-
tion of a mixture of enantiopure aldol adducts having an appro-
priate structure available for a subsequent diastereoselective
elimination reaction and a sequential reaction with the result-
ing aldol adducts which takes place to divergently afford syn-
and anti-propionate aldol adducts (Scheme 2). In order to real-
ize this sequential strategy, it is necessary to find a highly dias-
tereoselective transformation reaction which permits a diver-
gent elimination of the X substituent at the α-position of enan-
possibility.7
We briefly demonstrate here the feasibility of Guindon’s
methodology for stereocontrol through a radical reaction to
propionate aldol adducts.7 The stereochemical outcome of re-
actions involving an acyclic radical is rationalized by using the
transition state models: A, B, and C in Scheme 3. These transi-
tion state models, which are predictive of the predominant
product, take into consideration both steric and electronic fac-
tors. In A to the anti product, the opposition of the ester and
electronegative OR2 groups should reduce intramolecular elec-
trostatic repulsions. In addition, an electronic effect is in-
volved in the stabilization of an electron-poor radical by hyper-
conjugation with the best σ-donor substituent (R1). In B to the
syn product, 1,3-allylic strain appears to be a serious destabi-
lizing factor. Guindon concluded that C is the preferred transi-
tion state model to give the syn product; on the basis of the as-
sumption that the radical reaction proceeds through an early
transition state, both models A and C bear a marked similarity
to ground state conformer of the radical. The conclusion
reached was that the ratio of anti- and syn-products would be
dictated by the energy difference between transition states A
and C. As shown in Scheme 4, more effective stereocontrolled
systems have since been reported by Guindon;7 the reaction
which proceeds via an exocyclic radical gives, predominantly,
the anti product and the reaction, which can be chelation-con-
trolled with a variety of Lewis acids, shows a significant im-
provement in syn-selectivity.
Thus, the two required reactions, an enantioselective aldol
reaction and a diastereoselective radical reaction, appear to be
feasible to achieve our strategy to divergently prepare enan-
tiopure syn- and anti-propionate aldol adducts, as depicted in
Scheme 3. A rationale for the stereochemical outcome of
highly diastereoselective hydrogen-transfer reactions.
Scheme 4. More effectively stereocontrolled systems.

S. Kiyooka et al.
Bull. Chem. Soc. Jpn., 74, No. 8 (2001) 1487
Scheme 5.
Scheme 2. As mentioned above, we are already aware that a
large substituent at the β-position of ketene silyl acetals is re-
sponsible for the high enantioselectivity (approaching > 98%
ee) in the asymmetric aldol reaction in the presence of chiral
1,3,2-oxazaborolidin-5-one (S)-1, prepared in situ from L-Ts-
Val and BH3•THF.4k Based on this strategy, 2-bromo-2-meth-
ylketene ethyl silyl acetal 2 (bp 73–74 °C/0.1 mmHg, E : Z = 1
: 2) was chosen as an appropriate silyl nucleophile for this re-
action sequence. The bromo substituent in 2 has dual roles:
the first is to provide a suitable steric bulkiness of the silyl nu-
cleophile leading to very high enantioselectivity in the aldol
condensation process; the second is that it has promise as a
group which can be readily eliminated from the resulting inter-
mediates in the radical reduction process. The reaction of a va-
riety of aldehydes with the silyl nucleophile 2 proceeded in
good yields under the standard conditions (CH2Cl2, −78 °C,
15 h) in the presence of a stoichiometric amount of chiral bo-
rane (S)-1 (Scheme 5). A very high level of enantioselectivity
was observed.4u The considerably high diastereoselection ob-
served is particularly intriguing and appears to be beyond the
scope derived from the general features of the asymmetric al-
dol reaction which proceeds independent of the geometry of
the silyl nucleophiles used.4r,u
A systematic investigation of the influence of β-hydroxy
protecting groups on diastereoselection was carried for the rad-
ical reduction of ethyl 2-bromo-2-methyl-3-phenyl-3-(protect-
ed hyroxy)propionates with Bu3SnH and Et3B7h and a methyl
protected hydroxy function was found to be suitable for high
anti-diastereoselection, compared with benzyl or isopropyl
protection. However, the protecting group is not necessarily
pertinent for the practical synthesis of polypropionates, be-
cause the procedures required for protection and deprotection
are not simple. An easily eliminated TMS group seems to be
the most convenient protecting group for such transformations,
in that the protection is adequate. The above information
prompted us to investigate the radical debromination reduction
using starting compounds which contained a TMS protecting
group. The TMS protected starting materials syn-4a–d were
prepared from the corresponding aldol adducts syn-3 with TM-
SOTf and 2,6-lutidine in CH2Cl2 in good yields (Scheme 6).
Radical debromination reactions for a variety of essentially
enantiopure aldols syn-4a–d were investigated under the con-
ditions of Bu3SnH and Et3B in toluene at −78 °C, followed by
deprotection of the TMS function with an acidic methanolic
solution to afford the corresponding propionate aldol adducts
5. Basically the observed diastereoselection involves the same
level of anti-selection in the case where 4a contains a phenyl
Scheme 6.
Table 1. Effect of C-3 Substituents in α-Bromo-β-TMS-oxy
Esters on Diastereoselection in Radical Debromination
Reaction
Entry
4a–f
Isolated yield/% syn/anti
ratiob)
1
2
3
4
4a
4d
4c
4d
5a
5b
5c
5d
80
81
88
90
1 : 7.7
1 : 2.7
6.6 : 1
6.2 : 1
a) Conditions: 2 equiv of Bu3SnH, 0.2 equiv of Et3B,
Toluene, −78 °C, overnight.
b) Determined by NMR analysis.
substituent on C-3, though the selectivity (1 : 7.7) is somewhat
higher than that (1 : 4) observed for a similar system involving
OTBS.7h A larger protecting group for the β-hydroxy function
would likely destabilize an appropriate transition state to the
anti-isomer. When R on C-3 has a methylene unit adjacent to
the hydroxy carbon, the diastereoselection was reversed to syn-
predominance (entries 3 and 4) which must have arisen from
the transfer of hydrogen to a transition state C (Scheme 3). A
conclusion on the effect of C-3 substituents, derived from Ta-
ble 1, is that considerably large substituents prefer anti-selec-
tion and that linear substituents result in syn-selection, which
is consistent with the known prediction that the ratio of anti-
and syn-products would be dictated by the energy difference

1488 Bull. Chem. Soc. Jpn., 74, No. 8 (2001)
Divergent Synthesis of syn- and anti-Propionate
between transition states A and C.7g Unfortunately a general
trend for diastereoselection in the debromination reaction of α-
bromo-β-siloxy esters was not observed.
the C–Br bond forces the substrate molecule into a transition
state conformation in which the top face of the radical system
is exposed to the delivery of a hydrogen atom, thus providing
access to syn-diastereoselection. In addition, the R group, α,
to the radical system plays an important role in bloking the
down face of the radical system, thereby discouraging an at-
tack by Bu3SnH from that position. When the steric branch of
the R group was increased, the syn selectivity was substantially
enhanced (entry 2). Thus, an effective access to the practical
synthesis of essentially enantiopure syn-propionate aldol ad-
ducts was achieved via the use of a chiral 1,3,2-oxazaboroli-
din-5-one-mediated aldol reaction coupled with a stereoselec-
tive radical reduction. The sequence of these reactions might
provide a powerful approach for constructing chiral acyclic
skeletons involving syn-propionate units, as found in the C1–
C7 segment of Monensin.8
We next shifted our attention to an approach to the prepara-
tion of essentially enantiopure 2, 3-anti-propionate aldol ad-
ducts, according to the strategy, shown in Scheme 2, where the
“exocyclic effect” could be used as a reliable methodology to-
ward the anti-diastereoselective radical reduction. From the
viewpoint of a practical synthesis, we planned to construct 2,
3-anti-propionate aldol adducts involving a 3, 5-syn- or anti-
diol subunit. Such an enantiopure segment constitutes a typi-
cal unit found in a variety of natural products including the
C16–C19 segment of scytophycin C,9 the C4–C7 segment of
discodermolide,10 the C32–C35 subunit of the synthetic inter-
mediate for rapamycin,11 and the C19–C23 segment of
ionomycin12 (Scheme 9). We previously reported that a 1, 3-
syn- or anti-diol can be stereoselectively prepared by applying
a chiral (S)- or (R)-1,3,2-oxazaborolidin-5-one-promoted
asymmetric aldol reaction to ketene silyl acetal 10. Guindon
has already reported that the anti-propionate aldol adducts can
be obtained from a radical reduction of the acetals of 2-methyl-
2-phenylseleno-3,5-dihydroxypentanoate derivatives with
Bu3SnH via non-chelation control with respect to an exocyclic
We then planned to deal with the most simple case, that is,
the absence of any protection for the β-hydroxy function of the
starting bromo esters as a practical synthesis, since this would
considerably shorten the number of reaction steps by eliminat-
ing the protection and deprotection steps. A preliminary trial
(Scheme 7) was conducted using a mixture of syn- and anti-3a
with Bu3SnH and catalytic Et3B to afford an unacceptable level
of syn : anti selectivity (1 : 1.7). However, a reaction with
Bu3SnH and Et3B in the presence of MgBr2•OEt2, (chelation
controlled conditions, reported by Guindon)7c gave predomi-
nantly syn-2-methyl-3-hydroxypropionate esters 5 (> 98% ee)
in 87% yield with a syn : anti selectivity (5 : 1). It is notewor-
thy that the free hydroxy function did not interfere with the
Lewis acid and did not prevent coordination; also, the radical
from the β-hydroxy ester acted as a bidentate ligand. Reac-
tions of a variety of aldol adducts were investigated following
the same procedure described above. Each reduction resulted
in a syn-diastereoselection of nearly the same level (moderate-
ly high), except for entry 2, where a 14-fold increase in syn-se-
lectivity was observed, in good overall yields with very high
enantioselectivity (Table 2). To account for the mechanism of
the chelation-controlled radical reduction, a transition-state
model can be proposed, as depicted in Scheme 8. First, the
chelation of the carbonyl and hydroxy moieties to a bidented
Lewis acid (MgBr2•OEt2) as well as a homolytic cleavage of
Scheme 7.
Table 2. Diastereoselection of Chelation Controlled-Radical
Debromination of α-Bromo-β-hydroxy Estersa)
Scheme 8. A transition state model of chelation controlled-
radical reduction.
Entry
3 (R)
Ph (3a)
(CH3)2CH (3b)
CH3CH2CH2 (3c)
PhCH2CH2 (3d)
Yield/%b)
syn/anti
% ee
(syn)
1
2
3
4
87 (5a)
83 (5b)
79 (5c)
80 (5d)
5 : 1
> 70 : 1
5 : 1
> 98
> 98
> 98
> 98
6 : 1
a) Reaction (0 °C) of non-protected bromo aldol adducts (a
mixture of syn- and anti-isomers: 3a; 7 : 1, 3b; 16 : 1, 3c;
10 : 1, 3d; 9 : 1) was carried out with 2 equiv of Bu3SnH
and 0.2 equiv of Et3B, and 5 equiv of MgBr2•OEt2 in
CH2Cl2. b) Isolated yield.
Scheme 9. Synthetic targets as enantiopure subunits.

S. Kiyooka et al.
Bull. Chem. Soc. Jpn., 74, No. 8 (2001) 1489
Scheme 10.
lidene acetal moiety in the radical reductions. A rational ex-
planation of the stereochemical outcome in the hydrogen trans-
fer reaction can be given by using a transition model which ef-
fectively involved the so-called “exocyclic effect” (Scheme
12).
The epimeric bromides 15a–d involving the enantiopure
1,3-diol unit required for the synthesis of the targets under con-
sideration were prepared as follows (Scheme 13). The reaction
of isobutanal with silyl nucleophile 10 proceeded smoothly in
the presence of a stoichiometric amount of (S)- or (R)-chiral
borane to afford the corresponding dithiolane aldol adducts in
an essentially enantiopure state. The sequential procedure for
desulfurization was carried out using Bu3SnH with AIBN,13 in-
stead of Ni2B-H2,4g to give the enantiopure aldol adducts 12a,b
in nearly quantitative yields. The TMS protection of the hy-
droxy group in 12a,b was achieved with TMSOTf and 2,6-luti-
dine, and the ethoxycarbonyl function in the resulting TMS
protected aldol adducts was directly converted to the corre-
sponding aldehydes to give 13a,b upon the treatment of
DIBAL at −78 °C. The second asymmetric aldol reaction of
13a,b, in this sequence to 15a–d, with silyl nucleophile 2 in
the presence of (S)- or (R)-chiral borane furnished aldol ad-
ducts 14a–d (100% de at C3 and C5) in good yields. The ste-
reochemistry at the stereogenic center (C3) is completely con-
trolled by only the stereochemistry of the chiral borane used,
“promoter (catalyst) control” on the enantioselective acyclic
stereoselection, so that each 3, 5-syn- or anti-diol unit is readi-
ly available in a pure state of the relative configuration. After
deprotection of the TMS function in 14a–d, the resulting diol
was subjected to acetalization to give 15a–d, which are
epimeric only at C2.
The resultig epimeric bromides 15a–d underwent a radical
debromination reaction in the presence of Bu3SnH and Et3B in
toluene, similar to the reaction applied to 7, to afford the corre-
sponding target compounds 16a–d in high yields (86–92%)
with excellent anti diastereoselectivity (33–36 : 1), as shown in
Table 3. The stereochemistry of the major product was con-
firmed by the NMR study using their lactone derivatives
(Scheme 14). By using silica-gel flash column chromatogra-
phy, the target molecules were easily purified to give the enan-
tiopure 2, 3-anti-propionate aldol adducts involving a 3, 5-syn-
or anti-diol subunit. Thus, we were able to achieve effective
access to the preparation of essentially enantiopure 2, 3-anti-
Scheme 11.
Scheme 12.
effect.7 The above information persuaded us to investigate the
following reactions: after acetalization of the diols derived
from successive aldol reactions of an aldehyde with ketene si-
lyl acetals, 10 and 2, the subunits above might be prepared by
debromination using an exocyclic effect controlled radical re-
duction.
A preliminary trial was conducted with the corresponding
epimeric bromide 7. Bromide 7 was prepared by the chiral
1,3,2-oxazaborolidin-5-one-promoted asymmetric aldol reac-
tion of 3-t-butyldimethylsiloxypropanal with 2 in a stoichio-
metric amount of chiral (R)-1,3,2-oxazaborolidin-5-one to give
the corresponding α-bromo aldol adduct 6 in 92% yield with
95% ee at C3, followed by treatment with PTSA and 2,2-
dimethoxypropane (88% yield), in a one pot reaction. The
bromide 7 was then subjected to a hydrogen transfer reaction
with Bu3SnH with a catalytic amount of Et3B in toluene at −
78 °C with stirring overnight. The product 6 was preferentially
obtained in a ratio of 23 : 1 in 95% yield (Scheme 10). The
relative configuration of the reduced products 8 was deduced
by correlating the 1H NMR spectral data. A more rigorous ste-
reochemical assignment was performed for the anti-reduction
product 8 by subjecting it to acidic methanolic conditions to
afford lactone 9 (Scheme 11). Lactone 9 shows a smaller H2–
H3 coupling constant (3.5 Hz), proving that 8 contains a 2, 3-
anti configuration. The anti selectivity was shown to be quite
high in the case of the α-bromo system bearing an isopropy-

1490 Bull. Chem. Soc. Jpn., 74, No. 8 (2001)
Divergent Synthesis of syn- and anti-Propionate
Scheme 13.
Scheme 14.
quintet; m, multiplet), coupling constant(s) in Hertz, number of
protons. 13C NMR spectra were measured at 100 MHz with a
JEOL JNM-LA 400 spectrometer. High performance liquid chro-
matography (HPLC) was done with a JASCO Model PU-980 liq-
uid chromatograph, using DAICEL chiral columns. Optical rota-
tions were determined with a JASCO DIP-370 digital polarimeter.
Merck silica-gel 60 (230–400 mesh) was used for flash column
chromatography and for thin-layer chromatography Merck silica-
gel 60 TLC aluminum sheets were used.
propionate aldol adducts involving a 3, 5-syn- or anti-diol sub-
unit, which is available for the enantioselective synthesis of ap-
propriate fragments of several complex macrolides, by using
chiral 1,3,2-oxazaborolidin-5-one-promoted asymmetric aldol
reactions coupled with an anti-diastereoselective radical reduc-
tion.
Conclusion
We report the successful divergent synthesis of essentially
enantiopure syn- and anti-propionate aldol adducts by a new
methodology which effectively merged two reactions: the
enantioselective aldol reaction developed by Kiyooka4 and the
diastereoselective radical debromination developed by Guin-
don.7
2-Bromo-1-ethoxy-1-trimethylsiloxypropene (2).
To a
stirred solution of LDA (58.3 mmol) in THF (35 mL) (prepared in
situ) at −78 °C was added dropwise ethyl 2-bromopropionate (50
mmol) over 5 min. After 15 min of stirring at the same tempera-
ture, TMSCl (83.3 mmol) was added over 10 min. The reaction
mixture was allowed to warm to ambient temperature and THF
was removed with a rotary evaporator. The residue was mixed
with dry hexane and filtered. After removal of the solvent, the res-
idue was distilled under reduced pressure to give 2 (9.0 g, 71%, bp
73–74 °C/0.1 mmHg). IR (neat) 2962, 1655, 1251, 1219, 850
cm−1. Z-isomer: 1H NMR (CDCl3) δ 0.22 (s, 9H), 1.20 (t, J = 7.1
Experimental
General. Infrared spectra (IR) were determined with a JAS-
CO FT/IR-5300 Fourier transform infrared spectrometer. 1H
NMR spectra were determined by a JEOL JNM-LA 400 (a super-
conducting, 400 MHz, FT instrument) spectrometer. Chemical
shifts are expressed in ppm downfield from internal tetramethylsi-
1
Hz, 3H), 2.14 (s, 3H), 3.84 (q, J = 7.0 Hz, 2H). E-isomer: H
NMR (CDCl3) δ 0.19 (s, 9H), 1.22 (t, J = 7.1 Hz, 3H), 2.08 (s,
3H), 3.88 (q, J = 7.0 Hz, 2H).
1
lane. Significant H NMR data are tabulated in the following or-
der: multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; quin,
General Procedure for the Asymmetric Aldol Reaction in

S. Kiyooka et al.
Bull. Chem. Soc. Jpn., 74, No. 8 (2001) 1491
Table 3. High anti Selection in Radical Reduction Controlled by Exocyclic Effecta)
Diastereomeric ratiob)
Yield/%
86
Substrate
Product
anti/syn at C2–C3
33 : 1
36 : 1
35 : 1
33 : 1
92
90
88
a) All reactions were performed at a substrate concentration of 0.1 M using 2.0 equiv of
Bu3SnH with a catalytic amount of Et3B. b) Diastereomeric ratio was determined from
their 1H NMR spectra.
the Presence of a Stoichiometric Amount of a Chiral Borane
Complex. Ethyl (2S,3R)- and (2R,3R)-2-Bromo-3-hydroxy-2-
methyl-3-phenylpropionates (3a). Under an argon atmosphere,
to a stirred solution of (S)-1,3,2-oxazaborolidin-5-one (2.11 g,
7.78 mmol) in dry CH2Cl2 (10 mL) at 0 °C was added dropwise a
1 M solution of THF•BH3 in THF (7.01 ml, 7.01 mmol) over 5
min and the solution was stirred at 0 °C for 30 min. The resulting
solution was then allowed to warm to ambient temperature, and
was subsequently stirred for 30 min. The solution was then
cooled to −78 °C and PhCHO (0.68 mL, 6.56 mmol) in CH2Cl2 (1
mL) was added slowly over 5 min. Then, 2 was introduced over 5
min and the mixture was stirred for 15 h at −78 °C. The reaction
mixture was quenched by adding a buffer solution (10 mL, pH
6.8), extracted with CH2Cl2, and washed with sat. NaHCO3, fol-
lowed by sat. NaCl. After dried over anhydrous MgSO4 and evap-
orated, the residue was purified by flash column chromatography
(7% AcOEt in hexane) to separate 3 (1.45 g, 77.3% yield). Syn :
anti ratio was found on the basis of NMR data of the crude prod-
ucts. The physical data of 3a–d have been reported in ref. 4u. The
enantiomeric purity was determined by HPLC analysis. 3a: the
retention times of peak-1: 24.04 min and peak-2: 27.31 min [elu-
ent; hexane : 2-propanol (97.5 : 2.5), 0.5 mL min−1, Column:
CHIRALCEL OD]. 3b: the retention times of peak-1: 14.26 min
and peak-2: 18.87 min [eluent; hexane : 2-propanol (99.8 : 0.2), 1
mL min−1, Column : CHIRALPAK AD]. 3c: the retention times
of peak-1: 28.56 min and peak-2: 30.90 min [eluent; hexane : 2-
propanol (99 : 1), 0.5 mL min−1, Column: CHIRALPAK AD].
3d: the retention times of peak-1: 20.04 min and peak-2: 24.28
min [eluent; hexane : 2-propanol (97 : 3), 0.5 mL min−1, Column:
CHIRALCEL OD]. Found: C, 50.20; H, 5.31%. Calcd for
C12H15BrO3: C, 50.19; H, 5.27%.
min. Then, TMSOTf (0.66 mL, 3 mmol) was added and the mix-
ture was stirred for 2.5 h at the same temperature. The reaction
was quenched by a slow addition of sat. NaCl (10 mL). The reac-
tion mixture was extracted with ether, washed sat. NaCl, and dried
over anhydrous MgSO4. After the solvent was evaporated, the res-
idue was purified by flash column chromatography (1% AcOEt in
hexane) to afford 4a. Found: C, 50.10; H, 6.48%. Calcd for
C15H23BrO3Si: C, 50.14; H, 6.45%.
syn-4a (86% yield):[α]D20 −5.64 (c 1.24%, CHCl3). IR (neat)
2961, 1738 cm−1 1H NMR (CDCl3) δ 0.0 (s, 9H), 1.89 (t, J = 7.1
.
Hz, 3H), 1.69 (s, 3H), 4.10 (q, J = 7.1 Hz, 2H), 5.22 (s, 1H),
7.18–7.29 (m, 5H). 13C NMR (CDCl3) δ 0.0, 13.8, 21.8, 61.9,
67.0, 78.3, 127.7, 128.0, 128.1, 138.9, 170.2.
syn-4b (92% yield):[α]D21 −14.5 (c 2%, CHCl3). IR (neat)
2962, 1736 cm−1 1H NMR (CDCl3) δ 0.0 (s, 9H), 0.65 (d, J =
.
6.6 Hz, 3H), 0.69 (d, J = 6.8 Hz, 3H), 1.08 (t, J = 7.1 Hz, 3H),
1.35–1.41 (m, 1H), 1.59 (s, 3H), 3.91 (d, J = 2.9 Hz, 1H), 4.02
(dq, J = 1.7, 7.1 Hz, 2H). 13C NMR (CDCl3) δ 0.3, 13.3, 17.3,
21.6, 22.0, 31.0, 61.3, 67.0, 80.7, 170.5.
syn-4c (88% yield):[α]D21 +5.48 (c 1.64%, CHCl3). IR (neat)
2961, 1738 cm−1 1H NMR (CDCl3) δ 0.0 (s, 9H), 0.68 (t, J = 7.1
.
Hz, 3H), 0.92–1.07 (m, 2H), 1.09 (t, J = 7.1 Hz, 3H), 1.12–1.34
(m, 2H), 1.56 (s, 3H), 3.99 (dd, J = 1.7, 7.6 Hz, 1H), 4.02 (q, J =
7.1 Hz, 2H). 13C NMR (CDCl3) δ 0.9, 13.9, 13.9, 20.0, 21.5, 35.7,
61.9, 66.6, 77.3, 170.8.
syn-4d (90% yield):[α]D21 +21.25 (c 0.8%, CHCl3). IR (neat)
2957, 1736 cm−1 1H NMR (CDCl3) δ 0.0 (s, 9H), 1.01 (t, J = 7.1
.
Hz, 3H), 1.29–1.49 (m, 2H), 1.53 (s, 3H), 2.28 (ddd, J = 6.1, 10.7,
13.4 Hz, 1H), 2.58 (ddd, J = 5.1, 10.9, 13.7 Hz, 1H), 3.95 (q, J =
7.1 Hz, 2H), 4.02 (dd, J = 2.2, 9.5 Hz, 1H), 6.88–7.04 (m, 5H).
13C NMR (CDCl3) δ 13.8, 21.6, 33.3, 35.6, 61.9, 66.2, 77.5, 126.1,
128.3, 128.5, 141.4, 170.7.
Silylation Procedure to syn-4a–d. To a stirred solution of 3a
(574 mg, 2 mmol) in dry CH2Cl2 (16 mL) at room temperature
was added 2,6-lutidine (0.88 mL, 7.6 mmol) and stirred for 15
General Procedure for the Chelation Controlled-Radical
Reduction. Ethyl (2S,3S)- and (2R,3S)-3-Hydroxy-2-methyl-

1492 Bull. Chem. Soc. Jpn., 74, No. 8 (2001)
Divergent Synthesis of syn- and anti-Propionate
3-phenylpropionates (5a). Under an argon atmosphere Mg
turnings (180 mg, 7.1 mmol) were treated with 1,2-dibromoethane
(1.32 g, 7.1 mmol) at room temperature in dry diethyl ether (15
mL) until the reaction was complete, and then the reaction mixture
was evaporated. To a stirred suspension of the residue and 3a (287
mg, 1 mmol) in CH2Cl2 (10 mL) at 0 °C were added dropwise
Bu3SnH (1.34 mL, 5 mmol) and then Et3B (0.6 mL, 0.6 mmol) in
1/3 portions every 15 min; the mixture was then stirred for 2 h at
the same temperature. The reaction was quenched by adding m-
dinitrobenzene followed by pouring into an aqueous NaHCO3 so-
lution; the reaction was extracted with CH2Cl2, and washed with
sat.NaHCO3, followed by sat. NaCl. After being dried over anhy-
drous MgSO4 and evaporated, the residue was purified by flash
column chromatography (4% AcOEt in hexane) to afford 5a. The
syn : anti ratio was found on the basis of NMR data of the crude
products. The physical data of isolated isomers (5a–d) are report-
ed in Ref. 4l. The enantiomeric purity of each isomer was deter-
mined by HPLC analysis. 5a (87% yield): the retention times of
peak-1: 21.78 min and peak-2: 24.79 min [eluent; hexane : 2-pro-
panol (95 : 5), 0.5 mL min−1, Column: CHIRALCEL OD]. 5b
(83% yield): the retention times of peak-1: 11.88 min and peak-2:
62.2, 63.2, 73.6, 99.3, 169.8. Found: C, 44.75; H, 6.50%. Calcd
for C11H19BrO4: C, 44.76; H, 6.49%.
Ethyl (2S,3S)-3,5-Isopropylidenedioxy-2-methylpentanoate
(8). Aldol adduct 8 was prepared according to the usual proce-
dure using Bu3SnH with a catalytic amount of Et3B in toluene.
[α]D21 +20.1 (c 1.49%, CHCl3). 1H NMR (CDCl3) δ 1.10 (d, J =
7.1 Hz, 3H), 1.25 (t, J = 7.1 Hz, 3H), 1.34 (s, 3H), 1.43 (s, 3H),
1.46–1.50 (m, 1H), 1.59 (dABq, J = 12.0, 5.6 Hz, ∆ν = 20.8 Hz,
1H), 2.48 (dq, J = 7.1, 7.5 Hz, 1H), 3.86 (ddd, J = 1.7, 5.4, 11.7
Hz, 1H), 3.97 (dt, J = 3.2, 12.0 Hz, 1H), 4.08 (ddd, J = 2.4, 8.3,
11.2 Hz, 1H), 4.15 (q, J = 7.1 Hz, 2H). 13C NMR (CDCl3) δ 12.3,
14.2, 19.1, 27.9, 29.67, 45.6, 59.7, 60.2, 70.6, 98.4, 174.7. Found:
C, 60.12; H, 9.31%. Calcd for C11H20O4: C, 61.09; H, 9.32%.
Lactone 9. Under an argon atmosphere, PTSA (20 mg) was
added to a stirred solution of 8 (100 mg, 0.33 mmol) in dry MeOH
(5 mL) at room temperature. The resulting solution was stirred for
12 h at the same temperature. The reaction was quenched by the
slow addition of water (10 mL); then, the reaction mixture was ex-
tracted with ether, washed with sat. NaCl, and dried over anhy-
drous MgSO4. After the solvent was evaporated, the residue was
purified by flash column chromatography (20% AcOEt in hexane)
to afford the lactone 9 (80% yield). 1H NMR (CDCl3) δ 1.30 (d, J
= 7.1 Hz, 3H), 1.97–2.04 (m, 1H), 2.15–2.23 (m, 1H), 2.62 (dq, J
= 7.1, 3.7 Hz, 1H), 2.76 (d, J = 3.4 Hz, 1H), 4.24 (dt, J = 4.1, 4.2
Hz, 1H), 4.30 (dt, J = 5.8, 5.6 Hz, 1H), 4.56 (ddd, J = 4.9, 8.0,
11.2 Hz, 1H). 13C NMR (CDCl3) δ 11.9, 31.0, 41.3, 65.1, 67.2,
174.1.
14.84 min [eluent; hexane : 2-propanol (99.7 : 0.3), 1 mL min−1
,
Column: CHIRALCEL OD]. 5c (79% yield): the retention times
of peak-1: 15.19 min and peak-2: 17.52 min [eluent; hexane : 2-
propanol (99.8 : 0.2), 1 mL min−1, Column: CHIRALCEL OD].
5d (80% yield): the retention times of peak-1: 4.46 min and peak-
2: 10.14 min [eluent; hexane : 2-propanol (95 : 5), 1 mL min−1
Column: CHIRALCEL OD].
,
2-[Ethoxy(trimethylsiloxy)methylene]-1,3-dithiolane (10).
To a stirred solution (−78 °C) of LDA (125 mmol) in THF (250
mL) (prepared in situ) was added dropwise ethyl 2,2-(1,3-dithi-
olan-2-yl)acetate (100 mmol) for over 15 min. After 15 min of
stirring at the same temperature, TMSCl (300 mmol) was added
over 10 min. The reaction mixture was allowed to warm to ambi-
ent temperature, and stirred overnight. After THF was removed
with a rotary evaporator, the residue was mixed with dry hexane
and filtered. After removal of the solvent, the residue was distilled
under reduced pressure to give 10 (18.2 g, 72%, bp 108–110 °C/
Ethyl (2R,3S)-2-Bromo-5-(t-butyldimethylsiloxy)-3-hydro-
xy-2-methylpentanoate (6). This product was produced from
the reaction of 3-(t-butyldimethylsiloxy)propionaldehyde with 2
using a stoichiometric amount of (R)-1,3,2-oxazaborolidin-5-one
(87% yield). Syn isomer was purified from a mixture, of which
the syn : anti ratio was found to be 15 : 1 on the basis of NMR da-
ta.
[α]D24 −2.28 (c 3.4%, CHCl3). IR (neat) 3462, 1736 cm−1 1H
.
NMR (CDCl3) δ 0.0 (s, 6H), 0.82 (s, 9H), 1.23 (t, J = 7.1 Hz, 3H),
1.54–1.67 (m, 2H), 1.78 (s, 3H), 3.35 (d, J = 2.9 Hz, 1H), 3.71–
3.82 (m, 2H), 4.15 (dt, J = 2.9, 9.3 Hz, 1H), 4.18 (q, J = 7.1 Hz,
2H). 13C NMR (CDCl3) δ −5.5, −5.5, 13.8, 18.2, 23.0, 25.8,
34.6, 61.2, 62.2, 66.6, 74.6, 170.51. The enantiomeric purity of
syn-isomer was determined to be 95% ee by an HPLC analysis.
The retention times of peak-1 was 24.46 min and peak-2 was
0.1 mmHg). IR (neat) 2962, 1655, 1251, 1219, 850 cm−1
.
1H
NMR (CDCl3, 400 MHz) δ 0.0 (s, 9H), 1.02 (t, J = 7.1 Hz, 3H),
3.02–3.59 (m, 4H), 3.60 (q, J = 7.1 Hz, 2H).
Ethyl
(3R)-2,2-(Ethylenedithio)-3-hydroxy-4-methylpen-
tanoate (11a). This product was produced from the reaction of
isobutyraldehyde with 10 using a stoichiometric amount of (S)-
1,3,2-oxazaborolidin-5-one (75% yield). [α]D26 −44.6 (c 1.03%,
CHCl3). 1H NMR (CDCl3) δ 0.95 (d, J = 6.8 Hz, 3H), 1.0 (d, J =
6.6 Hz, 3H), 1.31 (t, J = 7.1 Hz, 3H), 1.84 (octet, J = 6.6 Hz, 1H),
3.30 (d, J = 6.8 Hz, 1H), 3.25–3.40 (m, 4H), 3.95 (dd, J = 6.1,
6.4 Hz, 1H), 4.24 (q, J = 7.1 Hz, 2H). 13C NMR (CDCl3) δ 13.9,
18.4, 20.7, 33.1, 39.7, 40.0, 62.8, 76.2, 78.8, 171.6.
22.38 min [eluent; hexane : 2-propanol (99.8 : 0.2), 0.5 mL min−1
,
Column: CHIRALPAK OD-H]. Found: C, 45.56; H, 7.85%. Cal-
cd for C14H29BrO4Si: C, 45.52; H, 7.91%.
Ethyl (2R,3S)-2-Bromo-3,5-isopropylidenedioxy-2-methyl-
pentanoate (7). Under an argon atmosphere, to a stirred solu-
tion of 6 (370 mg, 1 mmol) in dry MeOH (5 mL) at a room tem-
perature was added PTSA (20 mg); the mixture was stirred for 30
min. Then, 2,2-dimethoxypropane (5.0 mL, 40.9 mmol) was add-
ed and the mixture was further stirred for 30 min. The reaction
was quenched by the slow addition of water (10 mL). The result-
ing mixture was extracted with ether, washed with sat. NaCl, and
dried over anhydrous MgSO4. After the solvent was evaporated,
the residue was purified by flash column chromatography (8%
AcOEt in hexane) to afford 7 (88% yield). [α]D21 +28.4 (c 2.03%,
CHCl3). 1H NMR (CDCl3) δ 1.31 (t, J = 7.1 Hz, 3H), 1.42 (s,
3H), 1.47 (s, 3H), 1.45–1.51 (m, 1H), 1.75 (dABq, J = 12.0, 5.4
Hz, ∆ν = 21.2 Hz, 1H), 1.84 (s, 3H), 3.86 (ddd, J = 2.2, 5.6, 12.0
Hz, 1H), 3.96 (dt, J = 3.2, 12.0 Hz, 1H), 4.31 (dd, J = 11.5, 2.7
Hz, 1H). 13C NMR (CDCl3) δ 13.9, 19.2, 23.1, 26.5, 29.5, 59.4,
Ethyl
(3S)-2,2-(Ethylenedithio)-3-hydroxy-4-methylpen-
tanoate (11b). This product was produced from the reaction
with 10 using a stoichiometric amount of (R)-1,3,2-oxazaboroli-
din-5-one (71% yield). [α]D26 +45.0 (c 1.2%, CHCl3).
General Procedure for the Desulfurization. Ethyl (3S)-3-
Hydroxy-4-methylpentanoate (12a). To a stirred solution of
11a (765 mg, 3 mmol) in dry benzene (5 mL) at room temperature
were added Bu3SnH (1.76 mL, 6 mmol) and AIBN (150 mg). The
resulting mixture was refluxed for 1.5 h at 80 °C, allowed to cool,
evaporated, and purified by flash column chromatography (5%
AcOEt in hexane) to afford 12a (95% yield). [α]D26 −28.54 (c
1.15%, CHCl3). 1H NMR (CDCl3) δ 0.85 (d, J = 6.8 Hz, 3H),
0.88 (d, J = 6.8 Hz, 3H), 1.21 (t, J = 7.1 Hz, 3H), 1.60–1.70 (m,

S. Kiyooka et al.
Bull. Chem. Soc. Jpn., 74, No. 8 (2001) 1493
1H), 2.33 (dd, J = 10.0, 15.6 Hz, 1H), 2.42 (dd, J = 2.8, 16.4 Hz,
1H), 2.85 (d, J = 4.0 Hz, 1H), 3.68–3.73 (m, 1H), 4.10 (q, J = 7.1
Hz, 2H). 13C NMR (CDCl3) δ 14.1, 17.7, 18.3, 33.1, 38.4, 60.6,
72.6, 173.4.
Ethyl (2R,3S,5R)-2-Bromo-3-hydroxy-2,6-dimethyl-5-tri-
methylsiloxyheptanoate (14d). From the aldol reaction be-
tween 13d and 2 using a stoichiometric amount of (R)-1,3,2-ox-
azaborolidin-5-one, 14d was prepared (71% yield). [α]D26 +8.0 (c
3%, CHCl3). 1H NMR (CDCl3) δ 0.0 (s, 9H), 0.67 (d, J = 6.8 Hz,
3H), 0.73 (d, J = 6.8 Hz, 3H), 1.15 (t, J = 7.1 Hz, 3H), 1.37–1.49
(m, 2H), 1.62–1.70 (m, 1H), 1.70 (s, 3H), 3.34 (s, 1H), 3.64–3.68
(m, 1H), 3.98–4.03 (m, 1H), 4.07–4.17 (m, 2H). 13C NMR
(CDCl3) δ 0.0, 13.5, 16.7, 17.0, 22.3, 32.4, 33.8, 61.7, 65.7, 74.9,
76.4, 170.0.
Ethyl (3R)-3-Hydroxy-4-methylpentanoate (12b).
Aldol
adduct 12b was prepared from 11b (91% yield). [α]D26 +27.91 (c
1.57%, CHCl3).
(3S)-4-Methyl-3-trimethylsiloxypentanal (13a).
To a
stirred solution of 12a (300 mg, 1.87 mmol) in dry CH2Cl2 (10
mL) at room temperature was added 2,6-lutidine (0.86 mL, 7.48
mmol); the mixture was stirred for 15 min. Then, TMSOTf (0.61
mL, 2.8 mmol) was added and the mixture was stirred for 2.5 h at
the same temperature. The reaction was quenched by the slow ad-
dition of sat. NaCl (10 mL); the reaction mixture was then extract-
ed with ether, washed with sat. NaCl, and dried over anhydrous
MgSO4. After the solvent was evaporated, the residue was puri-
fied by flash column chromatography (1% AcOEt in hexane) to af-
ford TMS-protected product (92% yield). Under an argon atmo-
sphere, to a stirred solution of the TMS-protected product (254
mg, 1.2 mmol) in dry CH2Cl2 (10 mL) at −78 °C was added drop-
wise a 1 M solution of DIBAL in toluene (1.32 mL, 1.32 mmol)
over 30 min; the mixture was stirred at the same temperature for 2
h. The reaction was quenched by the slow addition of water (5
mL), extracted with ether, washed with sat. NaCl, dried over anhy-
drous MgSO4, evaporated and purified by flash column chroma-
tography (4% AcOEt in hexane) to afford 13a (84% yield). 1H
NMR (CDCl3) δ 0.0 (s, 9H), 0.77 (d, J = 6.8 Hz, 3H), 0.79 (d, J =
6.8 Hz, 3H), 1.59–1.67 (m, 1H), 2.24–2.48 (m, 2H), 3.87–3.92 (m,
1H), 9.69 (s, 1H).
Ethyl
(2R,3S,5S)-2-Bromo-3,5-isopropylidenedioxy-2,6-
dimethyl heptanoate (15a). Under an argon atmosphere, to a
stirred solution of 14a (80 mg, 0.21 mmol) in dry MeOH (5 mL)
at room temperature was added citric acid (200 mg); the mixture
was stirred for 15 min. Then, 2,2-dimethoxypropane (2.5 mL,
20.5 mmol) was added and further stirred for 30 min. The result-
ing solution was evaporated, followed by the addition of dry ace-
tone (10 mL) and then 2,2-dimethoxypropane (2.5 mL, 20.5
mmol), and stirred for 30 min. The reaction was quenched by the
slow addition of water (10 mL), extracted with ether, washed with
sat. NaCl, and dried over anhydrous MgSO4. After the solvent
was evaporated, the residue was purified by flash column chroma-
tography (4% AcOEt in hexane) to afford 15a (82% yield). [α]D26
−21.69 (c 1.06%, CHCl3). 1H NMR (CDCl3) δ 0.86 (d, J = 6.6
Hz, 3H), 0.92 (d, J = 6.6 Hz, 3H), 1.30 (t, J = 7.1 Hz, 3H), 1.34
(s, 3H), 1.36 (s, 3H), 1.56–1.67 (m, 2H), 1.72–1.79 (m, 1H), 1.82
(s, 3H), 3.39 (ddd, J = 7.3, 7.6, 10.0 Hz, 1H), 4.20 (dd, J = 6.7,
9.5 Hz, 1H), 4.22–4.29 (m, 2H). 13C NMR (CDCl3) δ 13.9, 17.6,
18.6, 23.5, 24.1, 24.2, 32.3, 32.8, 62.2, 64.1, 71.5, 71.9, 101.0,
169.0. Found: C, 49.77; H, 7.34%. Calcd for C14H25BrO4: C,
49.86; H, 7.47%.
(3R)-4-Methyl-3-trimethylsiloxypentanal (13b). Aldehyde
13b was prepared from 12b according to the procedure of 13a.
Ethyl
(2R,3S,5S)-2-Bromo-3-hydroxy-2,6-dimethyl-5-tri-
Ethyl
(2R,3R,5R)-2-Bromo-3,5-isopropylidenedioxy-2,6-
methylsiloxyheptanoate (14a). From an aldol reaction between
13a and 2 using a stoichiometric amount of (R)-1,3,2-oxazaboroli-
din-5-one, 14a was prepared (75% yield). Found: C, 45.62; H,
7.89%. Calcd for C14H29BrO4Si: C, 45.52; H, 7.91%. [α]D26 −9.37
(c 1.6%, CHCl3). 1H NMR (CDCl3) δ 0.0 (s, 9H), 0.72 (d, J = 7.1
Hz, 3H), 0.74 (d, J = 6.8 Hz, 3H), 1.16 (t, J = 7.1 Hz, 3H), 1.31–
1.43 (m, 2H), 1.61 (octet, J = 6.6 Hz, 1H), 1.70 (s, 3H), 3.11 (s,
1H), 3.59 (dt, J = 3.2, 6.6 Hz, 1H), 4.03 (br.d, J = 9.8 Hz, 1H),
4.07–4.17 (m, 2H). 13C NMR (CDCl3) δ 0.0, 13.5, 17.7, 18.1,
22.7, 33.1, 34.6, 61.9, 67.5, 72.2, 74.2, 170.3.
dimethyl heptanoate (15b). [α]D26 −15.85 (c 0.82%, CHCl3).
1H NMR (CDCl3) δ 0.86 (d, J = 6.8 Hz, 3H), 0.92 (d, J = 6.8 Hz,
3H), 1.30 (t, J = 7.1 Hz, 3H), 1.34 (s, 3H), 1.36 (s, 3H), 1.56–1.67
(m, 2H), 1.76 (ddd, J = 5.6, 9.5, 12.4 Hz, 1H), 1.82 (s, 3H), 3.40
(ddd, J = 6.1, 7.3, 9.8 Hz, 1H), 4.20 (dd, J = 6.6, 9.8 Hz, 1H),
4.21–4.31 (m, 2H). 13C NMR (CDCl3) δ 13.9, 17.6, 18.6, 23.5,
24.1, 24.2, 32.3, 32.8, 62.2, 64.1, 71.5, 71.9, 101.0, 169.9.
Ethyl
(2R,3R,5S)-2-Bromo-3,5-isopropylidenedioxy-2,6-
dimethyl heptanoate (15c). [α]D26 −23.12 (c 1.6%, CHCl3). 1H
NMR (CDCl3) δ 0.79 (d, J = 6.8 Hz, 3H), 0.84 (d, J = 6.8 Hz,
3H), 1.23 (t, J = 7.1 Hz, 3H), 1.33 (s, 3H), 1.36 (s, 3H), 1.18–1.33
(m, 1H), 1.39–1.43 (m, 1H), 1.49–1.59 (m, 1H), 1.76 (s, 3H), 3.41
(ddd, J = 2.2, 6.6, 11.2 Hz, 1H), 4.13–4.22 (m, 3H). 13C NMR
(CDCl3) δ 13.9, 17.7, 18.3, 19.6, 23.1, 28.8, 29.9, 33.1, 62.1, 63.5,
73.6, 74.0, 99.1, 169.9.
Ethyl (2R,3R,5R)-2-Bromo-3-hydroxy-2,6-dimethyl-5-tri-
methylsiloxyheptanoate (14b). From an aldol reaction between
13b and 2 using a stoichiometric amount of (S)-1,3,2-oxazaboroli-
din-5-one, 14b was prepared (78% yield). [α]D26 +7.5 (c 2%,
1
CHCl3). H NMR (CDCl3) δ 0.0 (s, 9H), 0.72 (d, J = 6.8 Hz, 3H),
0.74 (d, J = 6.8 Hz, 3H), 1.16 (t, J = 7.1 Hz, 3H), 1.31–1.43 (m,
2H), 1.61 (octet, J = 6.6 Hz, 1H), 1.70 (s, 3H), 3.09 (d, J = 3.4
Hz, 1H), 3.59 (ddd, J = 2.9, 5.9, 8.6 Hz, 1H), 4.03 (dt, J = 2.9,
10.0 Hz, 1H), 4.07–4.17 (m, 2H). 13C NMR (CDCl3) δ 0.0, 13.5,
17.7, 18.1, 22.7, 33.1, 34.6, 61.9, 67.5, 72.2, 74.2, 170.3.
Ethyl
(2R,3S,5R)-2-Bromo-3,5-isopropylidenedioxy-2,6-
dimethyl heptanoate (15d). [a]D26 +21.05 (c 1.9%, CHCl3). 1H
NMR (CDCl3) δ 0.86 (d, J = 6.8 Hz, 3H), 0.91 (d, J = 6.8 Hz,
3H), 1.30 (t, J = 7.1 Hz, 3H), 1.35–1.43 (m, 1H), 1.41 (s, 3H),
1.43 (s, 3H), 1.46–1.50 (m, 1H), 1.62 (octet, J = 6.8 Hz, 1H), 1.83
(s, 3H), 3.50 (ddd, J = 2.2, 6.6, 11.2 Hz, 1H), 4.23–4.24 (m, 1H),
4.27 (dq, J = 2.2, 7.1 Hz, 2H). 13C NMR (CDCl3) δ 13.9, 17.7,
18.2, 19.6, 23.1, 28.8, 29.9, 32.9, 62.0, 63.5, 73.6, 73.9, 99.1,
169.9.
Ethyl (2R,3R,5S)-2-Bromo-3-hydroxy-2,6-dimethyl-5-tri-
methylsiloxyheptanoate (14c). From an aldol reaction between
13c and 2 using a stoichiometric amount of (S)-1,3,2-oxazaboroli-
din-5-one, 14c was prepared (71% yield). [α]D26 −4.37 (c 1.6%,
1
CHCl3). H NMR (CDCl3) δ 0.0 (s, 9H), 0.66 (d, J = 7.1 Hz, 3H),
Ethyl (2S,3S,5S)-3,5-Isopropylidenedioxy-2,6-dimethylhep-
tanoate (16a). Under an argon atmosphere, to a stirred solution
of 15a in dry toluene (10 mL) at −78 °C was added dropwise
Bu3SnH (0.12 mL, 0.48 mmol), followed by Et3B (0.1 mL, 0.07
mmol); the mixture was stirred for 2 h at the same temperature.
The reaction was quenched by adding water (5 mL), extracted
0.73 (d, J = 6.8 Hz, 3H), 1.15 (t, J = 7.1 Hz, 3H), 1.35–1.40 (m,
2H), 1.60–1.70 (m, 1H), 1.70 (s, 3H), 3.34 (d, J = 2.2 Hz, 1H),
3.66 (dt, J = 3.9, 6.2 Hz, 1H), 4.0 (dt, J = 2.2, 5.6 Hz, 1H), 4.09
(dq, J = 2.4, 7.1 Hz, 2H). 13C NMR (CDCl3) δ 0.0, 13.5, 16.7,
17.0, 22.3, 32.4, 33.8, 61.7, 65.7, 75.0, 76.4, 170.0.

1494 Bull. Chem. Soc. Jpn., 74, No. 8 (2001)
Divergent Synthesis of syn- and anti-Propionate
with MeCN, and washed with sat. NaHCO3, followed by sat.
NaCl. The solution was dried over anhydrous MgSO4 and after
filtration the solvent was evaporated. The residue was purified by
flash column chromatography (2.5% AcOEt in hexane) to afford
16a (86% yield). Syn : anti ratio was determined on the basis of
NMR (CDCl3) δ 12.6, 17.5, 17.6, 32.5, 32.8, 41.4, 67.5, 80.5,
173.9.
This work was supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Science, Sports and
Culture.
NMR data. [α]D24 −37.09 (c 0.62%, CHCl3). IR (neat) 1739 cm−1
.
1H NMR (CDCl3) δ 0.86 (d, J = 6.6 Hz, 3H), 0.92 (d, J = 6.6 Hz,
3H), 1.10 (d, J = 7.1 Hz, 3H), 1.25 (t, J = 7.1 Hz, 3H), 1.29 (s,
3H), 1.30 (s, 3H), 1.52–1.69 (m, 3H), 2.28 (dq, J = 7.1, 7.1 Hz,
1H), 3.41 (dt, J = 7.1, 9.3 Hz, 1H), 3.93 (dt, J = 5.8, 9.5 Hz, 1H),
4.14 (m, 2H). 13C NMR (CDCl3) δ 12.5, 14.2, 17.6, 18.7, 24.1,
24.2, 32.9, 33.9, 45.6, 60.2, 68.5, 71.6, 100.5, 174.8. Found: C,
64.98; H, 10.21%. Calcd for C14H26O4: C, 65.09; H, 10.14%.
References
1
Enantioselective syntheses of syn-propionate aldol ad-
ducts: a) C. H. Heathcock, M. C. Pirrung, C. T. Buse, J. P. Hagen,
S. D.Young, and J. E. Sohn, J. Am. Chem. Soc., 101, 7077 (1979).
b) S. Masamune, W. Choy, F. A. J. Kerdesky, and B. Imperiali, J.
Am. Chem. Soc., 103, 1566 (1981). c) D. A. Evans and L. R.
McGee, J. Am. Chem. Soc., 103, 2876 (1981). d) D. A. Evans, J.
Bartroli, and T. L. Shih, J. Am. Chem. Soc., 103, 2127 (1981). e)
C. L. Hsiao, L. Liu, and M. J. Miller, J. Org. Chem., 52, 2201
(1987). f) E. J. Corey, R. Imwinkelried, S. Pikul, and Y. B. Xiang,
J. Am. Chem. Soc., 111, 5493 (1989). g) W. Oppolzer, J. Blagg, I.
Rodriguez, and E. Walther, J. Am. Chem. Soc., 112, 2767 (1990).
h) S. Kobayashi, H. Uchiro,Y. Fujishita, I. Shiina, and T.
Mukaiyama, J. Am. Chem. Soc., 113, 4247 (1991). i) W. Oppolzer
and P. Lienard, Tetrahedron Lett., 34, 4321 (1993). j) T. H. Yan,
H.-C. Lee, and C.-W. Tan, Tetrahedron Lett., 34, 3559 (1993). k)
S. G. Davis, G. J.-M. Doisneau, J. C. Prodger, and H. J. Sanganee,
Tetrahedron Lett., 35, 2373 (1994). l) J. B. Goh, B. R. Lagu, J.
Wurster, and D. C. Liotta, Tetrahedron Lett., 35, 6029 (1994). m)
R. K. Boeckman, A. T. Johnson, and R. A. Musselman, Tetrahe-
dron Lett., 35, 8521 (1994). n) T. Miyake, M. Seki, Y. Nakamura,
and H. Ohmizu, Tetrahedron Lett., 37, 3129 (1996). Enantioselec-
tive syntheses of anti-propionate aldol adducts: a) A. I. Meyers
and Y. Yamamoto, Tetrahedron, 40, 2309 (1984). b) C. Gennari,
A. Bernardi, L. Colombo, and C. Scolastico, J. Am. Chem. Soc.,
107, 5812 (1985). c) K. Narasaka and T. Miwa, Chem. Lett.,
1985, 1217. d) W. Oppolzer, Tetrahedron, 43, 1969 (1987). e) A.
G. Myers and K. L. Widdowson, J. Am. Chem. Soc., 112, 9672
(1990). f) E. J. Corey and S. S. Kim, J. Am. Chem. Soc., 112,
4976 (1990). g) H. Danda, M. M. Hansen, and C. H. Heathcock,
J. Org. Chem., 55, 173 (1990). h) M. A. Walker and C. H.
Heathcock, J. Org. Chem., 56, 5747 (1990). i) W. Oppolzer, C.
Strarkemann, I. Rodrigenz, and G. Bernardinelli, Tetrahedron
Lett., 32, 61 (1991). Reactions in the above literatures are some-
times stereodivergent by modifying the reaction conditions.
Ethyl
heptanoate (16b). [α]D24 +37.0 (c 0.62%, CHCl3). IR (neat)
1738 cm−1 1H NMR (CDCl3) δ 0.86 (d, J = 6.6 Hz, 3H), 0.92 (d,
(2R,3R,5R)-3,5-Isopropylidenedioxy-2,6-dimethyl-
.
J = 6.6 Hz, 3H), 1.10 (d, J = 6.8 Hz, 3H), 1.25 (t, J = 7.1 Hz,
3H), 1.29 (s, 3H), 1.30 (s, 3H), 1.52–1.69 (m, 3H), 2.48 (dq, J =
7.1, 7.3 Hz, 1H), 3.41 (dt, J = 6.4, 9.0 Hz, 1H), 3.94 (dt, J = 6.1,
9.3 Hz, 1H), 4.14 (m, 2H). 13C NMR (CDCl3) δ 12.5, 14.2, 17.6,
18.7, 24.1, 24.2, 32.9, 33.9, 45.6, 60.2, 68.4, 71.5, 100.5, 174.7.
Ethyl (2R,3R,5S)-3,5-Isopropylidenedioxy-2,6-dimethylhep-
tanoate (16c). [α]D24 −14.2 (c 0.42%, CHCl3). IR (neat) 1739
cm−1 1H NMR (CDCl3) δ 0.87 (d, J = 6.8 Hz, 3H), 0.91 (d, J =
.
6.6 Hz, 3H), 1.10 (d, J = 7.1 Hz, 3H), 1.12 (dt, J = 11.7, 12.0 Hz,
1H), 1.25 (t, J = 7.1 Hz, 3H), 1.34 (s, 3H), 1.39 (s, 3H), 1.49 (dt, J
= 2.2, 12.4 Hz, 1H), 1.58–1.66 (m, 1H), 2.48 (dq, J = 7.3, 7.5 Hz,
1H), 3.50 (ddd, J = 2.2, 6.6, 11.5 Hz, 1H), 4.01 (ddd, J = 2.4, 8.3,
11.5 Hz, 1H), 4.15 (q, J = 7.1 Hz, 2H). 13C NMR (CDCl3) δ 12.3,
14.2, 17.6, 18.3, 19.6, 30.0, 30.1, 33.0, 45.6, 60.2, 70.9, 73.7,
98.3, 174.8.
Ethyl (2S,3S,5R)-3,5-Isopropylidenedioxy-2,6-dimethylhep-
tanoate (16d). [α]D24 +13.9 (c 0.62%, CHCl3). IR (neat) 1738
cm−1 1H NMR (CDCl3) δ 0.87 (d, J = 6.8 Hz, 3H), 0.91 (d, J =
.
6.0 Hz, 3H), 1.10 (d, J = 7.1 Hz, 3H), 1.12 (dt, J = 11.7, 12.0 Hz,
1H), 1.25 (t, J = 7.1 Hz, 3H), 1.34 (s, 3H), 1.39 (s, 3H), 1.49 (dt, J
= 2.2, 12.4 Hz, 1H), 1.58–1.66 (m, 1H), 2.48 (dq, J = 7.1, 7.3 Hz,
1H), 3.50 (ddd, J = 2.1, 6.6, 11.2 Hz, 1H), 4.01 (ddd, J = 2.4, 8.3,
11.5 Hz, 1H), 4.15 (q, J = 7.1 Hz, 2H). 13C NMR (CDCl3) δ 12.3,
14.2, 17.6, 18.3, 19.6, 30.0, 30.1, 33.0, 45.6, 60.2, 70.9, 73.7,
98.3, 174.8.
Lactone 17. Under an argon atmosphere, to a stirred solution
of 16a (30 mg, 0.11 mmol) in dry MeOH (3 mL) at room temper-
ature was added PTSA (10 mg); the mixture was stirred for 12 h at
the same temperature. The reaction was quenched by the slow ad-
dition of water (5 mL), extracted with ether, washed with sat.
NaCl, and dried over anhydrous MgSO4. After the solvent was
evaporated, the residue was purified by flash column chromatogra-
phy (20% AcOEt in hexane) to afford lactone (89% yield). 1H
NMR (CDCl3) δ 0.98 (d, J = 6.8 Hz, 3H), 1.02 (d, J = 6.8 Hz,
3H), 1.29 (d, J = 7.1 Hz, 3H), 1.70 (ddd, J = 3.9, 11.5, 15.1 Hz,
1H), 1.79 (d, J = 3.9 Hz, 1H), 1.86–1.95 (m, 1H), 2.35 (ddd, J =
4.2, 8.0, 14.9 Hz, 1H), 2.69 (dq, J = 4.4, 7.1 Hz, 1H), 3.97 (ddd, J
= 4.2, 6.1, 11.0 Hz, 1H), 4.29 (dd, J = 3.6, 7.6 Hz, 1H). 13C
NMR (CDCl3) δ 10.9, 17.8, 17.8, 32.2, 34.7, 40.2, 67.7, 80.3,
173.9.
2
K. Ishihara, “Chiral B(ꢀ) Lewis Acids,” in “Lewis Acids in
Organic Synthesis,” ed by H. Yamamoto, Wiley-VCH, Weinheim
(2000), Chap. 5, pp. 135–189.
3
R. E. Gawley and J. Aubé, “Principles of Asymmetric Syn-
thesis,” Pergamon, Amsterdam (1996). See: a) K.Furuta, T.
Maruyama, and H. Yamamoto, Synlett. 1991, 439. b) References
in Ref. 4. c) E. R. Parmee, Y. Hong, O. Tempkin, and S.
Masamune, Tetrahedron Lett., 33, 1729 (1992).
4
a) S.-i. Kiyooka, Y. Kaneko, M. Komura, H. Matsuo, and
M. Nakano, J. Org. Chem., 56, 2276 (1991). b) S.-i. Kiyooka, Y.
Kaneko, and K. Kume, Tetrahedron Lett., 33, 4927 (1992). c) S.-i.
Kiyooka, Y. Kido, and Y. Kaneko, Tetrahedron Lett., 35, 5243
(1994). d) Y. Kaneko, T. Matsuo, and S.-i. Kiyooka, Tetrahedron
Lett., 35, 4107 (1994). e) Y. Kaneko and S.-i. Kiyooka, Mem. Fac.
Sci. Kochi Univ. Ser (c), 15, 9 (1994). f) S.-i. Kiyooka, Y. Kaneko,
Y. Harada, and T. Matsuo, Tetrahedron Lett., 36, 2821 (1995). g)
S.-i. Kiyooka and M. A. Hena, Tetrahedron: Asymmetry, 7, 2181
(1996). h) S.-i. Kiyooka, H. Kira, and M. A. Hena, Tetrahedron
Lactone 18. From 16c, lactone 18 was obtained (83% yield).
IR (neat) 1707 cm−1 1H NMR (CDCl3) δ 0.97 (d, J = 7.1 Hz,
.
3H), 0.99 (d, J = 7.1 Hz, 3H), 1.33 (d, J = 7.3 Hz, 3H), 1.78 (dd,
J = 12.4, 13.7 Hz, 1H), 1.86–1.94 (m, 1H), 1.94 (d, J = 3.2 Hz,
1H), 2.04 (ddd, J = 3.6, 3.9, 13.9 Hz, 1H), 2.50 (dq, J = 3.2, 7.1
Hz, 1H), 4.21 (s, 1H), 4.55 (ddd, J = 3.7, 5.6, 12.2 Hz, 1H). 13C

S. Kiyooka et al.
Bull. Chem. Soc. Jpn., 74, No. 8 (2001) 1495
Lett., 37, 2597 (1996). i) S.-i. Kiyooka, T. Yamaguchi, H. Maeda,
H. Kira, M. A. Hena, and M. Horiike, Tetrahedron Lett., 38, 3553
(1997). j) S.-i. Kiyooka and H. Maeda, Tetrahedron: Asymmetry,
8, 3371 (1997). k) S.-i. Kiyooka, Rev. Heteroatom Chem., 17, 245
(1997). l) M. A. Hena, S. Terauchi, C.-S. Kim, M. Horiike, and
S.-i. Kiyooka, Tetrahedron: Asymmetry, 9, 1883 (1998). m) S.-i.
Kiyooka, H. Maeda, M. A. Hena, M. Uchida, C.-S. Kim, and M.
Horiike, Tetrahedron Lett., 39, 8287 (1998). n) M. A. Hena, C.-S.
Kim, M. Horiike, and S.-i. Kiyooka, Tetrahedron Lett., 40, 1161
(1999). o) S.-i. Kiyooka and M. A. Hena, J. Org. Chem., 64, 5511
(1999). p) S.-i. Kiyooka, M. A. Hena, and F. Goto, Tetrahedron:
Asymmetry, 10, 2871 (1999). q) S.-i. Kiyooka, K. A. Shahid, and
M. A. Hena, Tetrahedron Lett., 40, 6447 (1999). r) S.-i. Kiyooka
and K. A. Shahid, Tetrahedron Lett., 41, 2633 (2000). s) S.-i.
Kiyooka, K. Goh, Y. Nakamura, H. Takesue, and M. A. Hena, Tet-
rahedron Lett., 41, 6599 (2000). t) S.-i. Kiyooka, M. A. Hena, T.
Yabukami, K. Murai, and F. Goto, Tetrahedron Lett., 41, 7511
(2000). u) S.-i. Kiyooka and K. A. Shahid, Tetrahedron: Asymme-
try, 11, 1537 (2000).
Yoakim, D. Hall, R. Lemieux, and B. Simoneau, Tetrahedron
Lett., 32, 27 (1991). c) Y. Guindon, J.-F. Lavalée, M. Llinas-
Brunet, G. Horner, and J. Rancourt, J. Am. Chem. Soc., 113, 9701
(1991). d) Y. Guindon, C. Yoakim, V. Gorys, W. W. Ogilvie, D.
Delorme, J. Renaud, G. Robinson, J.-F. Lavalée, A. Slassi, G.
Jung, J. Rancourt, K. Durkin, and D. Liotta, J. Org. Chem., 59,
1166 (1994). e) Y. Guindon, A. Slassi, J. Rancourt, G. Bantle, M.
Benchequroun, L. Murtagh, L. É. Ghiro, and G. Jung, J. Org.
Chem., 60, 288 (1995). f) Y. Guindon, Z. Liu, and G. Jung, J. Am.
Chem. Soc., 119, 9289 (1997). g) Y. Guindon, A. M. Faucher, E.
Bourque, V. Caron, G. Jung, and S. R. Landry, J. Org. Chem., 62,
9276 (1997). h) Y. Guindon and J. Rancourt, J. Org. Chem., 63,
6554 (1998). i) Y. Guindon, G. Jung, B Guérin, and W. W.
Ogilvie, Synlett, 1998, 213.
8
K. C. Nicolaou and E. J. Sorensen, “Classic in Total Syn-
thesis,” VCH, Weinheim (1996).
W. R. Roushang and G. J. Dilley, Tetrahedron Lett., 40,
9
4955 (1999).
10 D. E. Evans, D. P. Halstead, and B. D. Allison, Tetrahedron
Lett., 40, 4461 (1999).
11 K. C. Nicolaou, T. K. Chakraborty, A. D. Piscopio, N.
Minowa, and P. Bertinato, J. Am. Chem. Soc., 115, 4419 (1993).
12 A. Montana, F. García, and P. M. Grima, Tetrahedron Lett.,
40, 1375 (1999).
5
6
Preliminary results have been reported in Refs. 4q and 4r.
A. Ghosez, B. Giese, T. Göbel, and H. Zipse, “Formation
of C-H Bonds via Radical Reactions,” in “Stereoselective Synthe-
sis,” ed by G. Helmchen, R. W. Hoffmann, J. Mulzer, and E.
Schaumann, Houben-Weyl: Stuttgart (1996), D. 2.2, Vol. 7, E 21.
7
a) Y. Guindon, C. Yoakim, R. Lemieux, L. Boisvert, D.
13 C. G. Gutierrez, R. A. Stringham, T. Nitasaka, and K. G.
Glasscock, J. Org. Chem., 45, 3393 (1980).
Delorme, and J.-F. Lavalée, Tetrahedron Lett., 31, 2845 (1990). b)
Y. Guindon, J.-F. Lavalée, L. Boisvert, C. Chabot, D. Delorme, C.