Reversed stereochemical control in the presence of CeCl3 and TiCl4 in the Lewis acid mediated reduction of α-alkyl-β-keto esters by metal hydrides. A general methodology for the diastereoselective synthesis of syn- and anti-α-alkyl-β

Article abstract of DOI:10.1021/jo9821574

The Lewis acid-mediated reduction of α-alkyl-β-keto esters has been shown to proceed by different stereochemical control depending on the nature of the metal atom. Strongly chelating TiCl₄ led to the syn isomer in high diastereomeric excess in

Full text of DOI:10.1021/jo9821574

1986
J . Org. Chem. 1999, 64, 1986-1992
Rever sed Ster eoch em ica l Con tr ol in th e P r esen ce of CeCl3 a n d
TiCl4 in th e Lew is Acid Med ia ted Red u ction of r-Alk yl-â-k eto
Ester s by Meta l Hyd r id es. A Gen er a l Meth od ology for th e
Dia ster eoselective Syn th esis of syn - a n d a n ti-r-Alk yl-â-h yd r oxy
Ester s
Enrico Marcantoni,* Sara Alessandrini, and Marco Malavolta
Dipartimento di Scienze Chimiche, via S. Agostino 1, I-62032 Camerino (MC), Italy
Giuseppe Bartoli,* Maria Cristina Bellucci, and Letizia Sambri
Dipartimento di Chimica Organica “A. Mangini”, v.le Risorgimento 4, I-40136 Bologna, Italy
Renato Dalpozzo
Dipartimento di Chimica, Universita` della Calabria, I-87030 Arcavacata di Rende, Cosenza, Italy
Received October 27, 1998
The Lewis acid-mediated reduction of R-alkyl-â-keto esters has been shown to proceed by different
stereochemical control depending on the nature of the metal atom. Strongly chelating TiCl4 led to
the syn isomer in high diastereomeric excess in noncoordinating solvents (CH2Cl2) at -78 °C with
BH3‚py as reducing agent, while nonchelating CeCl3 gave a high excess of the anti isomer in
coordinating solvents (THF) at the same temperature with lithium triethylborohydride (LiEt3BH)
as reducing agent. The methodology has been successfully utilized for obtaining important syn-
and anti-R-alkyl-â-hydroxy esters with high diastereoselectivity.
In tr od u ction
tion of various metal enolates, which has been success-
fully employed for such a purpose.8
The stereochemical course in the reduction of the
carbonyl group of ketones with an adjacent chiral center
to the corresponding alcohol has long been studied.1
Particularly, interest has been focused on the diastereo-
selective reduction of R-substituted â-functionalized car-
bonyl compounds.2 Among these, the chemical3 or enzy-
mological4 stereoselective reduction of R-substituted â-keto
esters has been extensively studied in recent years.
Accordingly, two strategies (chelation and nonchelation)
have been developed that enabled the achievement of an
opposite sense of diastereoselectivity by the appropriate
choice of reducing systems. These methods have been
successfully applied to a variety of natural product
syntheses5 including â-lactams6 and â-lactone antibiot-
ics.7 Thus, the stereoselective reduction of the corre-
sponding R-substituted â-keto ester to highly function-
alized R-substituted â-hydroxy carboxylate represents an
attractive alternative to stereoselective aldol condensa-
In the course of our program to develop new synthetic
Lewis acid-mediated reductions of â-functionalized car-
bonyl compounds with asymmetric R-carbon in their
racemic form, we have found that the nature of the Lewis
acid is instrumental in determining the stereochemical
outcome of these reactions.9 Herein we wish to report that
the correct choice of hydrides, solvent, and especially
Lewis acid allows us to obtain high stereoselective
synthesis of R-substituted â-hydroxy esters. Our proce-
dure represents a simple and effective alternative pro-
cedure for the highly diastereoselective synthesis of syn-
or anti-R-alkyl-â-hydroxy esters through the reduction
of the corresponding â-keto esters with BH3‚py or lithium
triethylborohydride (LiEt3BH) in the presence of TiCl4
or dry CeCl3, respectively (Scheme 1).
(1) Greeves, N. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 7, pp 1-24.
(2) Oishi, T.; Nakata, T. Acc. Chem. Res. 1984, 17, 338.
(3) (a) Sato, T.; Nishio, M.; Otera, J . Synlett 1995, 965. (b) Taniguchi,
M.; Fujii, H.; Oshima, K.; Utimoto, K. Tetrahedron 1993, 49, 11169.
(c) Fujita, M.; Hiyama, T. J . Org. Chem. 1988, 53, 5405. (d) Fujita,
M.; Hiyama, T. J . Am. Chem. Soc. 1985, 107, 8294. (e) Ito, Y.;
Yamaguchi, M. Tetrahedron Lett. 1982, 24, 5385. (f) Sorrell, T. N.;
Pearlman, P. S. Tetrahedron Lett. 1980, 21, 3963.
(7) (a) Mead, K. T.; Park, M. Tetrahedron Lett. 1995, 36, 1205. (b)
Brown, H. C.; Kulkarni, S. V.; Racherla, U. S. J . Org. Chem. 1994, 59,
365. (c) Hanessian, S.; Tehim, A.; Chen, P. J . Org. Chem. 1993, 58,
7768. (d) Thompson, K. L.; Chang, M. N.; Chiang, Y.-C. P.; Yang, S.
S.; Chabala, J . C.; Arison, B. H.; Greenspan, M. D.; Hauf, D. P.;
Yudkovitz, J . Tetrahedron Lett. 1991, 32, 3337. (e) Chiang, Y. P.; Yang,
S. S.; Heck, J . V.; Chabala, J . C.; Chang, M. N. J . Org. Chem. 1989,
54, 5708. (f) Marshall, J . A.; Lebreton, J . J . Am. Chem. Soc. 1988, 110,
2925.
(4) (a) Dauchet, S.; Bigot, C.; Buisson, D.; Azerad, R. Tetrahedron:
Asymmetry 1997, 8, 1735. (b) Shieh, W.-R.; Sih, C. J . Tetrahedron:
Asymmetry 1993, 4, 1259. (c) J ian-Xiu, G.; Gno-Qiang, L. Tetrahedron
1993, 49, 5805. (d) Servi, S. Synthesis 1990, 1. (e) Nakamura, K.; Miyai,
T.; Nagar, A.; Oka, S.; Ohno, A. Bull. Chem. Soc. J pn. 1989, 62, 1179.
(5) Bartlett, P. A. Tetrahedron 1980, 36, 3.
(6) (a) Shirai, F.; Nakai, T. Chem. Lett. 1989, 445. (b) Wild, H.; Kant,
J .; Walker, D. G.; Ojima, I.; Ternansky, R. J .; Morin, J . M., J r.; Georg,
G. I.; Ravikumar, V. T. In The Organic Chemistry of â-lactams; Georg,
G. I., Ed.; VCH Publishers: New York, 1993; and references therein.
(8) (a) Bal, B.; Buse, C. T.; Smith, K.; Heathcock, C. H. Org. Synth.
1985, 63, 89. (b) Montgomery, S. H.; Pirrung, M. C.; Heathcock, C. H.
Org. Synth. 1985, 63, 99. (c) Meyers. A. I.; Yamamoto, Y. Tetrahedron
1984, 40, 2309. (d) Heathcock, C. H.; Pirrung, M. C.; Montgomery, S.
H.; Lampe, J . Tetrahedron 1981, 37, 4087.
(9) (a) Bartoli, G.; Bosco, M.; Cingolani, S.; Marcantoni, E.; Sambri,
L. J . Org. Chem. 1998, 63, 3624. (b) Bartoli, G.; Bosco, M.; Dalpozzo,
R.; Marcantoni, E.; Sambri, L. Chem. Eur. J . 1997, 3, 1941. (c) Bartoli,
G.; Bosco, M.; Marcantoni, E.; Sambri, L. Tetrahedron Lett. 1996, 35,
241.
10.1021/jo9821574 CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/04/1999

Reversed Stereochemical Control of CeCl3 and TiCl4
J . Org. Chem., Vol. 64, No. 6, 1999 1987
Ta ble 1. Ster eoselective Red u ction of â-Keto Ester s (1) to â-Hyd r oxy Ester s w ith BH3‚p y in th e P r esen ce of TiCl4 in
CH2Cl2 a t -78 °C
Sch em e 1a
Sch em e 2
a
Key: (a) TiCl4 in CH2Cl2 at -78 °C, then BH3‚py; (b) CeCl3 in
THF at -78 °C, then LiEt3BH.
Resu lts a n d Discu ssion
The additions of nucleophiles to ketones are promoted
by coordination of a Lewis acid to the carbonyl group,
enhancing the electrophilicity of this moiety,10 and bo-
rane-Lewis base complexes are finding an increasing
role for this purpose.11 It is also known that the reduction
of R-substituted â-keto esters produces the syn diastere-
omer when a Lewis acid, which is able to coordinate to
two additional ligands, is used.12 We found that the
reduction of R-alkyl-â-keto esters 1 in dichloromethane
at low temperature (-78 °C) with BH3‚py13 (1.5 equiv)
in the presence of TiCl4 (1.5 equiv) led to the prevalent
formation of syn-R-alkyl-â-hydroxy esters (syn-2) (Table
1). The observed stereochemical course suggests a preva-
lent TiCl4-induced chelation-controlled addition,14 and the
1-TiCl4 complex can be represented by an equilibrium
between the A and B conformations (Scheme 2). These
species, involving 1 equiv of TiCl4, should be the reactive
intermediate, as suggested by the observation that ad-
dition of a second equivalent of Lewis acid has no
influence on the diastereoselectivity (Table 1, entry 2).
The high diastereoselectivity observed in this reduction
is probably a consequence of the generally strong ability
of titanium(IV) compounds to engage in chelation-
controlled carbonyl addition.15 It is obvious that every
increase in bulkiness of R1, shifting the conformational
equilibrium toward the A conformation, increases the
syn/ anti ratio. In fact, the cylindrical symmetry of the
alkynyl group in R1 position reduces the unfavorable
steric interaction between the R1 substituent and the
oxygen atom of the carbonyl group in conformation B
(Table 1, entry 5).
(10) Sweeney, J . B. In Comprehensive Organic Functional Group
Transformation; Katritzky, A. R., Meth-cohn, O., Rees, C. W., Eds.;
Elsevier Science Ltd.: Oxford, 1995; Vol. 2, pp 52-82.
(11) (a) Pelter, A.; Smith, K.; Brown, H. C. Borane Reagents;
Academic Press: London, 1988. (b) Lane, C. F. Aldrichim. Acta 1973,
6, 51.
(12) Nakata, T.; Oishi, T. Tetrahedron Lett. 1980, 21, 1641.
(13) Sarko, C. R.; Collibee, S. E.; Knorr, A. L.; DiMare, M. J . Org.
Chem. 1996, 61, 868.
(14) Reviews on chelation- and nonchelation-controlled additions:
(a) Reetz, M. T. Angew. Chem., Int. Ed. Engl. 1984, 23, 556. (b) Reetz,
M. T. Acc. Chem. Res. 1993, 26, 462. (c) J onas, V.; Frenking, G.; Reetz,
M. T. Organometallics 1993, 12, 2110.
(15) Reetz, M. T. In Organometallics in Synthesis; Schlosser, M.,
Ed.; J ohn Wiley & Sons: Chichester, 1994; pp 195-282.

1988 J . Org. Chem., Vol. 64, No. 6, 1999
Marcantoni et al.
Ta ble 2. Ster eoselective Red u ction of â-Keto Ester s (1) to â-Hyd r oxy Ester s w ith LiEt3BH in th e P r esen ce of Dr y CeCl3
in THF a t -78 °C
Sch em e 3
The reduction of â-keto esters 1 with BH3‚py in the
presence of TiCl4 was studied in different solvents such
as CH2Cl2, THF, and Et2O. Dichloromethane turned out
to be a suitable solvent for the reduction, since the use
of a coordinating solvent, such as THF and Et2O, led to
the complete loss of diastereoselectivity. To control di-
astereoselectivity, the reductions were executed at low
temperature (-78 °C), as temperatures higher than -78
°C considerably lower both stereoselectivity and yields.
We also examined the reduction of R-substituted â-keto
esters 1 with hydrides in the presence of dry cerium
trichloride as the Lewis acid. We first tested the reduction
of 1a in dichloromethane with BH3‚py, a convenient
reducing agent because of its solubility in many organic
solvents, in the presence of dry CeCl3. While the reducing
system BH3‚py was highly efficient with TiCl4, no reac-
tion with CeCl3 was observed, and at -78 °C the starting
material was quantitatively recovered after 2 h. Pro-
longed reaction times and higher reaction temperatures
were also ineffective. A complex between cerium(III)
chloride and â-keto ester was formed, since the addition
of the substrate to a suspension of the cerous salt resulted
in a clear solution after about 1 h. Unfortunately, the
paramagnetism of cerous salt prevented useful NMR
information on the structure of this complex from being
chains favor the C over the D conformation, and the
hydride anion attacks the â-carbonyl at the opposite side
to the bulky carboalkoxy group selectively leading to the
anti isomer (Scheme 3). This explanation was further
supported by the fact that anti-diastereoselecitivity was
enhanced in the reduction of more bulky tert-butyl esters
(Table 2, entries 8 and 9). For this reason, we believe
obtained.16
The reduction of R-alkyl-â-keto esters 1 proceeded
successfully when a 1 M THF solution of LiEt3BH was
employed in the presence of dry CeCl3 and at low
temperature (-78 °C). The reaction gave the expected
R-alkyl-â-hydroxy esters 2 in almost quantitative yields,
but with reversed stereoselectivity with respect to the
that the cerium trichloride cannot participate in chelation
BH3‚py/TiCl4 system (Table 2). In this latter reducing
when a six-membered transition state is involved, even
though the CeCl3 is apparently a good Lewis acid and
system we observed that the important factor in the syn-
selective reduction was the intermediary formation of
its anhydrous form is highly hydroscopic.20
stable titanium chelates. Therefore, for the anti-reduc-
tion, it is obvious that the use of Lewis acids having
(17) (a) Che´rest, M.; Felkin, H.; Prudent, N. Tetrahedron Lett. 1968,
metals of high O-chelating ability are to be avoided.
These findings then strongly support an open-chain
mechanism, and the anti selectivity for the system LiEt3-
BH/CeCl3 may be explained by a Felkin-Ahn model17-19
controlled reaction. In fact, the sterically demanding alkyl
2199. (b) Anh, N. T.; Einsenstein, O. Nouv. J . Chim. 1977, 1, 61.
(18) (a) Bu¨rgi, H. B.; Dunitz, J . D.; Shefter, E. J . J . Am. Chem. Soc.
1973, 95, 5065. (b) Bu¨rgi, H. B.; Dunitz, J . D.; Lehn, J . M.; Wipff, G.
Tetrahedron 1974, 30, 1563.
(19) Bu¨rgi, H. B.; Lehn, J . M.; Wipff, G. J . Am. Chem. Soc. 1974,
96, 1956.
(20) The water molecule found in dry cerium(III) chloride seems to
have no effect, since the material was highly efficient without a large
excess of reducing agent. cf. Evans, W. J .; Feldaman, J . D.; Ziller, J .
W. J . Am. Chem. Soc. 1996, 118, 4581.
(16) Hubert-Pfalzgraf, L. G.; Machado, L.; Vaissermann, J . Polyhe-
dron 1996, 15, 546.

Reversed Stereochemical Control of CeCl3 and TiCl4
J . Org. Chem., Vol. 64, No. 6, 1999 1989
Ta ble 3. 1H NMR Ch em ica l Sh ifts for th e Ca r bin ol Reson a n ces of â-Hyd r oxy Ester s (2)
Moreover, since Greeves et al.21 have shown quite
clearly that cerium(III) can be chelated by 1,2-diols, we
have considered the possibility that the real reducing
agent of the ketone moiety is not the Et3BH- anion. It
may be, namely, that, since both LiEt3BH and CeCl3 are
required for efficient and diastereoselective reduction at
-78 °C, the hydride of boron can reduce the CeCl3 to
though an incomplete inversion to syn products was
observed, as it was found in the presence of the powerful
chelating TiCl4. This would confirm the poor chelating
ability of dry cerium(III) trichloride with these â-keto
esters 1.
Although several types of metal hydride complexes are
available for this reduction of R-alkyl-â-keto esters, the
lithium triethylborohydride was among the best. The
data clearly reveal that LiEt3BH is an exceptionally
powerful hydride reducing agent25 and that it exibits a
remarkable selectivity, a property normally considered
to be characteristic of relatively mild reducing agent, such
as lithium borohydride.26
1
CeHnCl3-n in a polar solvent such as THF. However, H
NMR measurements demostrated that, under these
conditions, there was no generation of cerium hydride
species via the transmetalation between cerium trihalide
and lithium triethylborohydride.22 In addition, Fukazawa
et al.23 have suggested that such hydrides species are
unlikely to be the reducing agent in the reduction
reactions with LiAlH4 in the presence of CeCl3; therefore,
from our results, a similar behavior seems to be plausible.
The dry cerium(III) chloride, then, is not able to chelate
with â-keto esters, but its presence is essential to obtain
high yields and diastereoselectivities. In fact, the reaction
of â-keto esters 1 with LiEt3BH alone24 was slower and
the yields were lower with respect to the same reaction
carried out with CeCl3. Reduction with short reaction
times could be obtained at 0 °C, but it is detrimental to
the diastereoselectivity and the yields, since an insoluble
polymer is formed. At 0 °C in the presence of CeCl3, the
same low yields and low diastereoselectivity are observed.
Cerium(III) trichloride is therefore essential in achieving
high yields and diastereoselectivity, since it allows the
reaction to be performed at low temperature. In these
studies, THF was found to be a suitable solvent for the
reaction, since the use of this coordinating solvent
ensures good yields and high diastereoselectivity. On the
other hand, the use of noncoordinating solvent such as
CH2Cl2 caused a decrease in diastereoselectivity, even
The assignment for syn and anti diastereomers is based
on the coupling constant between the R- and â-protons,
which is usually slightly larger for the anti than syn
isomer27 (Table 3). Moreover, the stereochemistry of syn
and anti diastereomers was fixed by using their 13C NMR
chemical shifts of carbinol carbon in comparison with the
reported data.28 It has also been showen that in the crude
products the syn/anti ratios were very similar or identical
to those of the purified â-hydroxy esters. Therefore, we
can reasonably assume that the observed syn/anti ratio
at the â-hydroxy ester stage reflects the diastereofacial
selectivity of the addition step.
Further, on extending our procedure to obtain both syn-
and anti-R-alkyl-â-hydroxy esters by reduction of the
corresponding â-keto esters, the stereoselectivity of the
reduction on â-keto esters 1 in the presence of other
functionalities was examined. The isolated ester function
is stable under our reaction conditions (Table 1, entry 2;
Table 2, entry 2), whereas, when the â-keto ester
substrate contains an acetal moiety (1g), the carbonyl
group can be selectively reduced with the reducing
system LiEt3BH/CeCl3 (Table 2, entry 7). On the other
hand, we observed that if the reducing system BH3‚py/
(21) Greeves, N.; Pease, J . E.; Bowden, M. C.; Brown, S. M.
Tetrahedron Lett. 1996, 37, 2675.
(22) Preparation of 1H NMR sample: A 5-mm NMR tube charged
with CeCl3‚7H2O (79.4 mg, 0.21 mmol) was heated (140 °C) in vacuo
(0.1 mmHg) for 2 h and charged with nitrogen. Then the tube was
charged with dry THF-d8 (0.75 mL) and frozen by liquid nitrogen. Next,
a THF-d8 solution of LiEt3BH (0.15 mL, 0.14 mmol) was added by
syringe onto the solidified THF. Then the mixture was frozen again,
and the tube was flame-sealed under vacuum. The 1H NMR spectrum
was measured after warming to -78 °C, it was referenced to residual
tetrahydrofuran (δ 3.58 ppm, δ 1.73 ppm), and a broad new peak did
not appear.
(25) Brown, H. C.; Kim, S. C.; Krishnamurthy, S. J . Org. Chem.
1980, 45, 1.
(26) Bartoli, G.; Bosco, M.; Marcantoni, E.; Sambri, L. Tetrahedron
Lett. 1996, 37, 7421.
(27) House, H. O.; Crumrine, D. S.; Teramishi, A. Y.; Olmstead, H.
D. J . Am. Chem. Soc. 1973, 95, 3310.
(28) (a) Frater, G.; Mu¨ller, U.; Gu¨nther, W. Tetrahedron 1984, 40,
1277. (b) Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, T.
C.; Sohn, J . E.; Lampe, J . J . Org. Chem. 1980, 45, 1066. (c) Mulzer, J .;
Zippel, M.; Brumtrup, G.; Segner, J .; Finke, J . Liebigs Ann. Chem.
1980, 1108. (d) Heathcock, C. H.; Pirrung, M. C.; Sohn, J . E. J . Org.
Chem. 1979, 44, 4294.
(23) Fukuzawa, S.; Fujinami, T.; Yamanchi, S.; Sakai, S. J . Chem.
Soc., Perkin Trans. 1 1986, 1929.
(24) Ito, Y.; Katsuki, T.; Yamaguchi, M. Tetrahedron Lett. 1985, 26,
4643.

1990 J . Org. Chem., Vol. 64, No. 6, 1999
Marcantoni et al.
Sch em e 4
Sch em e 5
TiCl4 was used, no selectivity was obtained, and the
acetal moiety was also cleaved29 (Table 1, entry 7).
Unfortunately, both reducing systems are unsuccessful
when the R-alkyl chain of the â-keto esters 1 contains
another isolated carbonyl moiety (1f). Indeed, the reduc-
tion of both carbonyl groups provides dihydroxycarboxy-
lates 3 (Scheme 4); however, no attempt to isolate this
intermediate was undertaken. Instead, treatment with
acid during workup directly afforded 3,6-disubstituted
δ-lactone 2f as mixture of diastereomers (Table 1, entry
6; Table 2 entry 6).
that complexation of the keto ester is unimportant in the
case of these reduction conditions. Finally, it is notewor-
thy that our methodology can be carried out under
standard procedures without any particular precaution
and provides easy access to wide varieties of â-hydroxy
carboxylates with high syn and anti diastereoselectivities.
Finally, we have also examined our reducing systems,
BH3‚py/TiCl4 and LiEt3BH/CeCl3, with R-substituted
carbonyl compounds that have no ability to chelate with
metals. Our choice was ketone 4 because the tert-butyl
group may be bulky enough to cause strong stereodiffer-
entation beteween E and F conformations (Scheme 5).
The high diastereoselectivity (syn-5/anti-5 ) 2:98 and
Exp er im en ta l Section
1
5:95 by H NMR30) can be qualitatively understood by
Gen er a l P r oced u r es. 1H NMR spectra were recorded at
200 MHz using residual CHCl3 (7.26 ppm) as reference. 13C
NMR spectra were recorded at 50 MHz with CDCl3 (77.05
ppm) as reference. Mass spectra were measured at an ionizing
voltage of 70 eV. All reactions were performed in flame-dried
glassware under a positive pressure of nitrogen and stirred
magnetically. Column and thin-layer chromatographies were
carried out on silica gel (230-400 mesh ASTM) with the
solvent system ethyl acetate in hexanes as eluent. Diastere-
omeric purity was determined by NMR analysis.
Ma ter ia ls. Commercial grade reagents and solvents were
used without further purification; otherwise, where necessary,
they were purified as recommended.31 â-Keto esters 1a and
1c were prepared by mono-C-alkylation of nonsubstituted
â-keto esters with sodium alkoxide.32 â-Keto esters 1d , 1e, and
1i were prepared by a heterogeneous alkylation of ethyl and
tert-butyl benzoyl acetate.33 â-Keto esters 1b, 1g, and 1h were
prepared from the corresponding bromides via C-alkylation of
nonsubstituted â-keto esters with NaH.34 â-Keto ester 1f was
prepared through a cerium(III) chloride-catalyzed Michael
reaction of the corresponding â-keto ester and methyl vinyl
an open-chain mechanism, and the anti selectivity may
explained by Felkin-Ahn’s model. With this substrate,
then, both TiCl4 and CeCl3 are not able to chelate, and
they gave a high excess of the anti isomer with BH3‚py
and LiEt3BH as reducing agent, respectively.
Con clu sion
In summary, we have developed a straightforward
approach for the synthesis both syn- and anti-R-alkyl-â-
hydroxy esters, which are key intermediates for the
stereoselective preparation of complex functionalized
molecules, by Lewis acid-mediated reduction of R-alkyl-
â-keto esters with metal hydrides. The nature of the
Lewis acid was found to be pivotal in determining the
outcome of these reductions. Strongly chelating TiCl4 led
largely to the syn diastereomer with BH3‚py as reducing
agent, while nonchelating CeCl3 gave an excess of the
anti diastereomer with LiEt3BH as reducing agent. This
would indicate that â-keto esters are probably just
inadequate ligands for dry CeCl3, and given the vast
excess of polar THF solvent over the substrate, it is likely
(31) Armarego, W. L. F.; Perrin, D. D. In Purification of Laboratory
Chemicals, 4th ed.; Butterworth Heinemann: Oxford, 1996.
(32) Martin, V. A.; Murray, D. H.; Pratt, N. E.; Zhao, Y.; Albizati,
K. F. J . Am. Chem. Soc. 1990, 112, 6965.
(29) Sammakia, T.; Smith, R. S. J . Org. Chem. 1992, 57, 2997.
(30) (a) Arjona, O.; Pe´rez-Ossorio, R.; Pe´rez-Rubalcaba, A.; Quiroga,
M. L. J . Chem. Soc., Perkin Trans. 2 1981, 597. (b) Alvarez-Ibarra, C.;
Arjona-Loraque, O.; Pe´rez-Rubalcaba, A.; Plumet, J . Rev. R. Acad.
Cienc. Exactas, Fis. Nat. Madrid 1979, 73.
(33) Queignes, R.; Kirschleger, B.; Lambert, F.; Aboutaj, M. Synth.
Commun. 1998, 18, 1213.
(34) (a) Kim, H.-O.; Carrol, B.; Lee, M. S. Synth. Commun. 1997,
27, 2505. (b) Reynolds, R. C.; Trask, T. W.; Sedwick, W. D. J . Org.
Chem. 1991, 56, 2391.

Reversed Stereochemical Control of CeCl3 and TiCl4
J . Org. Chem., Vol. 64, No. 6, 1999 1991
ketone as acceptor.35 The tert-butyl ketone 4 was obtained by
means of the tertiary group introduction via the Friedel-
Crafts alkylation of the corresponding trimethylsilyl enol ether
according to reported procedure.36 Other starting materials
were commercially available, while tert-butyl benzoyl acetate
was prepared according to the published procedure.37
70.37, 74.06, 82.05, 126.79, 128.35, 128.79, 141.90, 173.18; EI-
MS m/z 232 (M+), 204, 159, 115(100), 91. Anal. Calcd for
C14H16O3: C, 72.39; H, 6.94. Found: C, 72.36, H, 6.91.
3-[H yd r oxy(p h en yl)m et h yl]-6-m et h ylt et r a h yd r o-2H -
p yr a n -2-on e (2f). Flash column chromatography (30% EtOAc
in hexanes) afforded δ-lactone 2f as a diastereomeric mixture.
Our attempts to separate the diastereomers were unsuccess-
ful: IR (neat) 3418, 1725 cm-1; 1H NMR (CDCl3) δ 1.34 (d, 3H
J ) 6.30 Hz), 1.93-2.03(m, 3H), 2.38-2.43 (m, 1H), 2.94-2.97
(m, 1H), 3.05 (bs, 1H, OH), 4.30-4.51 (m, 1H), 5.48 (d, 1H, J
) 2.48 Hz), 5.54 (d, 1H, J ) 2.99 Hz), 7.26-7.85 (m, 5H); 13C
NMR (CDCl3) δ 21.05, 25.81, 25.89, 30.14, 30.86, 47.31, 47.64,
70.72, 72.05, 73.00, 73.08, 126.20, 127.57, 128.13, 128.75,
142.03, 142.36, 171.96. Anal. Calcd for C13H16O3: C, 70.89; H,
7.32; Found: C, 70.86; H, 7.30.
Gen er a l P r oced u r e for Ster eoselective Red u ction of
r-Alk yl-â-k eto Ester s 1 w ith BH3‚p y in th e P r esen ce of
TiCl4. A representative experiment is as follows. To a cold
(-78 °C) solution of â-keto esters 1 (1.0 mmol) in 10 mL of
dry CH2Cl2 was added TiCl4 (1.5 mmol, solution 1 M in CH2-
Cl2) to give immediately a clear solution, which was stirred
for 15 min at this temperature. The complex BH3‚py (1.5 mmol)
in 5 mL of CH2Cl2 was then added. After 15 min, 25 mL of 1
N HCl was added, and the reaction was warmed to room
temperature. The organic layer was separated, the aqueous
layer was washed with CH2Cl2, and the combined organics
were concentrated in vacuo. The resulting residue was parti-
tioned between Et2O and H2O. The etheral layer was washed
with water and brine, dried over Na2SO4, and concentrated
in vacuo. Flash column chromatography gave syn-R-alkyl-â-
hydroxy esters 238 contaminated by only a minor amount of
the anti diastereomer. Diastereomeric purity was determined
by NMR analysis, and reported yields listed in Table 1 are
based on material isolated by flash column chromatography
on silica gel using hexanes-EtOAc as the eluent system.
Eth yl (2R*,3S*)-2-ben zyl-3-h yd r oxybu ta n oa te (syn -2a ):
IR (neat) 3435, 1735 cm-1; 1H NMR (CDCl3) δ 1.09 (t, 3H, J )
7.24 Hz), 1.26 (d, 3H, J ) 6.43 Hz), 2.35 (bs, 1H, OH), 2.73-
2.79 (m, 1H), 2.98 (d, 2H, J ) 7.66 Hz), 4.13 (q, 2H, J ) 7.22
Hz), 4.30 (after addition of D2O, dq, 1H, J ) 6.42, 5.37 Hz),
7.16-7.28 (m, 5H); 13C NMR (CDCl3) δ 14.20, 25.15, 34.03,
52.05, 60.25, 68.33, 126.05, 128.16, 129.32, 139.23, 173.96; EI-
MS m/z 222 (M+), 204, 131 (100), 104, 91, 65, 51. Anal. Calcd
for C13H18O3: C, 70.24; H, 8.16. Found: C, 70.22; H, 8.12.
Eth yl (5R*,6S*)-5-eth oxyca r bon yl-6-h yd r oxyep ta n oa te
(syn -2b): IR (neat) 3495, 1732 cm-1; 1H NMR (CDCl3) δ 1.15-
1.30 (m, 9H), 1.56-1.78 (m, 4H), 2.20 (t, 2H, J ) 6.92 Hz),
2.51 (bs, 1H, 0H), 2.68-2.76 (m, 1H), 3.95 (t, 1H, J ) 5.82
Hz), 4.10-4.20 (m, 4H); 13C NMR (CDCl3) δ 14.39, 20.52, 21.57,
23.09, 26.96, 28.89, 34.19, 52.15, 60.79, 68.08, 173.56, 175.09;
EI-MS m/z 231, 202, 128(100), 111, 101, 81. Anal. Calcd for
Eth yl 2-Eth oxyca r bon yl-5-(2-h yd r oxyeth oxy)-1-p h en -
ylh exa n -1-ol (2g). The reaction of 1g with BH3‚py/TICl4 gave
2g as a diastereomeric mixture. Our attempts to separate the
diastereomers were unsuccessful: IR (neat) 3419, 1728 cm-1
;
1H NMR (CDCl3) δ 1.20 (d, 3H, J ) 6.22 Hz), 1.27 (t, 3H, J )
6.95 Hz), 1.51-1.74 (m, 4H), 3.06-3.15 (m, 1H), 3.50-3.69 (m,
5H), 4.14 (q, 2H, J ) 6.95 Hz), 4.35-4.49 (m, 1H), 4.53-4.65
(m, 1H), 6.02 (bs, 1H, OH), 7.10-7.40 (m, 5H); 13C NMR
(CDCl3) δ 14.49, 19.98, 20.05, 23.93, 24.35, 34.32, 34.68, 53.35,
53.81, 60.97, 61.04, 62.40, 70.01, 74.84, 74.95, 75.35, 76.41,
126.78, 126.85, 128.27, 128.77, 142.16, 142.25, 175.19. Anal.
Calcd for C16H26O5: C, 64.41; H, 8.78. Found: C, 64.38; H,
8.77.
ter t-Bu tyl (2R*,3S*)-2-ben zyl-3-h ydr oxybu tan oate (syn -
2h ): IR (neat) 3508, 1712 cm-1 1H NMR (CDCl3) δ 1.26 (d,
;
3H, J ) 6.38 Hz), 1.29 (s, 9H), 2.62-2.70 (m, 1H), 2.75 (bs,
1H, OH), 2.95 (dd, 2H, J ) 6.24 and 3.05 Hz), 4.00 (q, 2H, J
) 5.49 Hz), 7.17-7.26 (m, 5H); 13C NMR (CDCl3) δ 20.28,
27.80, 33.63, 54.75, 67.99, 81.07, 126.14, 128.19, 129.06,
139.23, 173.81; EI-MS m/z 232, 194, 131, 91(100). Anal. Calcd
for C15H22O3: C, 71.97; H, 8.86. Found: C, 71.93; H, 8.84.
ter t-Bu t yl (2R*,3R*)-2-a llyl-3-h yd r oxy-3-p h en ylp r o-
p a n oa te (syn -2i): IR (neat) 3444, 1724 cm-1; 1H NMR (CDCl3)
δ 1.27 (s, 9H), 2.42 (t, 2H, J ) 7.39 Hz), 2.64-2.74 (m, 1H),
3.01 (bs, 1H, OH), 4.84 (d, 1H, J ) 5.56 Hz), 4.96-5.09 (m,
2H), 5.68-5.77 (m, 1H), 7.25-7.34 (m, 5H); 13C NMR (CDCl3)
δ 27.88, 32.17, 53.24, 74.16, 81.18, 116.57, 126.57, 127.72,
128.21, 135.57, 141.52, 173.42; EI-MS m/z 206, 143, 107(100),
77, 41. Anal. Calcd for C16H22O3: C, 73.25; H, 8.45. Found: C,
73.21; H, 8.42.
C
12H20O5: C, 59.00; H, 8.25. Found: C, 58.98; H, 8.23.
E t h yl (2R*,3R*)-3-h yd r oxy-2-m et h yl-3-p h en ylp r o-
p a n oa te (syn -2c): IR (neat) 3478, 1715 cm-1; 1H NMR (CDCl3)
δ 1.13 (d, 3H, J ) 7.17 Hz), 1.19 (t, 3H, J ) 7.00 Hz), 2.74-
2.80 (m, 1H), 3.12 (bs, 1H, OH), 4.11 (q, 2H, J ) 7.04 Hz),
5.06 (d, 1H, J ) 4.43 Hz), 7.28-7.35 (m, 5H); 13C NMR (CDCl3)
δ 11.42, 14.55, 47.00, 61.18, 74.23, 126.53, 127.95, 128.70,
142.02, 176.25; EI-MS m/z 208 (M+), 163, 107, 102(100), 91,
65, 51. Anal. Calcd for C12H16O3: C, 69.27; H, 7.74. Found: C,
69.23; H, 7.70.
Gen er a l P r oced u r e for Ster eoselective Red u ction of
R-Alk yl-â-k eto Ester s 1 w ith LiEt3BH in th e P r esen ce of
Dr y CeCl3. A representative experiment is as follows. Finely
ground CeCl3‚7H2O (3.2 mmol) was dried by heating at 140
°C/0.1 Torr for 2 h,39 and then it was suspended in 10 mL of
dry THF and left to stir overnight at room temperature. At
this temperature, a solution of 1 (1.0 mmol) in 5 mL of THF
was added and left to stir for 1 h. Then it was cooled to -78
°C, and LiEt3BH (2.0 mmol, solution 1 M in THF) was added
by syringe. The reaction mixture was then left to stir until
TLC or GC indicated that no â-keto ester remained (Table 2).
The reaction mixture was quenched with diluted HCl (10%)
and extracted with Et2O. The organic layer was dried over
MgSO4, filtered, and evaporated to give anti-R-alkyl-â-hydroxy
esters 2 contaminated only by a minor amount of the syn
diastereomer. Diastereomeric purity was determined by NMR
analysis, and reported yields listed in Table 2 are based on
material isolated by flash column chromatography on silica
gel using hexanes-EtOAc as the eluent system.
Eth yl (2R*,3R*)-2-a llyl-3-h yd r oxy-3-p h en ylp r op a n oa te
(syn -2d ): IR (neat) 3460, 1728, cm-1; 1H NMR (CDCl3) δ 1.09
(t, 3H, J ) 7.00 Hz), 2.40-2.50 (m, 2H), 2.72-2.93 (m, 1H),
2.96 (bs, 1H,OH), 4.00 (q, 2H, J ) 7.02 Hz), 4.90 (d, 1H, J )
5.97 Hz), 4.95-5.10 (m, 2H), 5.62-5.86 (m, 1H), 7.20-7.40 (m,
5H); 13C NMR (CDCl3) δ 14.55, 32.26, 53.33, 61.01, 74.51,
117.26, 126.77, 128.77, 128.89, 135.88, 141.96, 174.59; EI-MS
m/z 234 (M+), 193, 143, 91, 79 (100), 77, 41. Anal. Calcd for
C
14H18O3: C, 71.77; H, 7.74. Found: C, 71.75; H, 7.69.
Eth yl (2R*,3R*)-3-h yd r oxy-3-p h en yl-2-p r op a r gylp r o-
p a n oa te (syn -2e): IR (neat) 3472, 1728 cm-1; 1H NMR (CDCl3)
δ 1.06 (t, 3H, J ) 7.20 Hz), 2.50-2.65 (m, 3H), 2.85-2.95 (m,
1H), 4.01(q, 2H, J ) 7.20 Hz), 4.90 (d, 1H, J ) 6.57 Hz), 7.20-
7.40 (m, 5H); 13C NMR (CDCl3) δ 14.49, 17.76, 52.90, 61.25,
Eth yl (2R*,3R*)-2-ben zyl-3-h yd r oxybu ta n oa te (a n ti-
2a ): IR (neat) 3435, 1728 cm-1; 1H NMR (CDCl3) δ 1.19 (t, 3H,
J ) 7.21 Hz), 1.30 (d, 3H, J ) 6.19 Hz), 1.90 (bs, 1H, OH),
3.05-3.20 (m, 1H), 3.39 (d, 2H, J ) 7.87 Hz), 4.05 (q, 1H, J )
6.30 Hz), 4.14 (q, 2H, J ) 7.19 Hz), 7.20-7.29 (m, 5H); 13C
NMR (CDCl3) δ 14.20, 25.36, 34.10, 52.45, 60.12, 68.71, 126.15,
128.19, 129.32, 138.75, 174.37; EI-MS m/z 205, 157, 91(100),
(35) Bartoli, G.; Bosco, M.; Bellucci, M. C.; Marcantoni, E.; Sambri,
L.; Torregiani, E. Eur. J . Org. Chem. 1999, 617.
(36) Reetz, M. T.; Chatziiosifidis, I.; Hubner, F.; Heimbach, H. Org.
React. 1984, 31, 101.
(37) Turner, J . A.; J acks, W. S. J . Org. Chem. 1989, 54, 4229.
(38) Descriptors R*,S* indicate that diastereomeric compounds are
obtained as racemates. We prefer this terminology to avoid the
ambiguities that could arise from syn-anti descriptors.
(39) Imamoto, T. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 1, pp 231-250.

1992 J . Org. Chem., Vol. 64, No. 6, 1999
Marcantoni et al.
65, 51. Anal. Calcd for C13H18O3: C, 70.24; H, 8.16. Found: C,
70.21; H, 8.13.
108.02, 126.20, 127.55, 128.13, 141.59, 174.95. Anal. Calcd for
C16H24O5: C, 64.84; H, 8.16. Found: C, 64.80; H, 8.15.
Eth yl (5R*,6R*)-5-eth oxyca r bon yl-6-h ydr oxyep ta n oa te
(a n ti-2b): IR (neat) 3456, 1738 cm-1; 1H NMR (CDCl3) δ 1.16-
1.31 (m, 9H), 1.56-1.75 (m, 4H), 2.30 (t, 2H, J ) 6.75 Hz),
2.60 (bs, 1H, OH), 2.68-2.76 (m, 1H), 3.90 (after addition of
D2O, t, 1H, J ) 6.22 Hz), 4.15 (q, 4H, J ) 7.08 Hz); 13C NMR
(CDCl3) δ 14.70, 14.75, 21.96, 23.13, 29.24, 34.51, 52.77, 60.81,
61.08, 68.75, 173.67, 175.59; EI-MS m/z 231, 202, 128 (100),
111, 101, 81. Anal. Calcd for C12H20O5: C, 59.00; H, 8.25.
Found: C, 58.96, H, 8.22.
Syn d ia ster eom er : IR (neat) 3436, 1728 cm-1 1H NMR
;
(CDCl3) δ 1.32 (t, 3H, J ) 7.00 Hz), 1.36 (s, 3H), 1.45-1.62
(m, 2H), 1.96 (t, 2H, J ) 7.60 Hz), 3.20-3.29 (m, 1H), 3.82-
3.95 (m, 4H), 4.26 (q, 2H, J ) 7.01 Hz), 4.95 (d, 1H, J ) 5.98
Hz), 6.10 (bs, 1H, OH), 7.18-7.35 (m, 5H); 13C NMR (CDCl3)
δ 14.25, 23.59, 24.19, 37.23, 52.92, 60.52, 64.86, 65.02, 72.98,
107.92, 126.70, 127.32, 128.15, 142.63, 175.23. Anal. Calcd for
C
16H24O5: C, 64.84; H, 8.16. Found: C, 64.82; H, 8.12.
ter t-Bu tyl (2R*,3R*)-2-ben zyl-3-h ydr oxybu tan oate (a n ti-
E t h yl
(2R*,3S*)-3-h yd r oxy-2-m et h yl-3-p h en ylp r o-
2h ): IR (neat) 3508, 1712 cm-1 1H NMR (CDCl3) δ 1.25 (d,
;
p a n oa te (a n ti-2c): IR (neat) 3458, 1716 cm-1
;
1H NMR
3H, J ) 7.37 Hz), 1.33 (s, 9H), 2.55-2.63 (m, 1H), 2.93 (d, 2H,
J ) 7.06 Hz), 2.88 (bs, 1H, OH), 3.85 (dq, 1H, J ) 7.34 and
6.44 Hz), 7.19-7.27 (m, 5H); 13C NMR (CDCl3) δ 21.71, 27.87,
35.60, 54.36, 67.72, 81.29, 126.26, 128.21, 129.06, 138.78,
174.22; EI-MS m/z 235, 194, 131(100), 103, 91. Anal. Calcd
for C15H22O3: C, 71.97; H, 8.86; Found C, 71.90; H, 8.84.
ter t-Bu t yl (2R*,3S*)-2-a llyl-3-h yd r oxy-3-p h en ylp r o-
(CDCl3) δ 1.01 (d, 3H, J ) 7.20 Hz), 1.15 (t, 3H, J ) 7.12 Hz),
2.75-2.83 (m, 1H), 2.96 (bs, 1H, OH), 4.16 (q, 2H, J ) 7.12
Hz), 4.70 (d, 1H, J ) 8.46 Hz), 7.31-7.38 (m, 5H); 13C NMR
(CDCl3) δ 14.62, 14.76, 47.62, 61.27, 76.91, 127.16, 128.31,
128.93, 142.09, 176.40; EI-MS m/z 208 (M+), 163, 107, 102-
(100), 91, 55. Anal. Calcd for C12H26O3: C, 69.27; H, 7.74.
Found: C, 69.25; H, 7.71.
p a n oa te (a n ti-2i): IR (neat) 3450, 1725 cm-1 1H NMR
;
Eth yl (2R*,3S*)-2-a llyl-3-h yd r oxy-3-p h en ylp r op a n oa te
(a n ti-2d ): IR (neat) 3460, 1726 cm-1; 1H NMR (CDCl3) δ 1.18
(t, 3H, J ) 7.00 Hz), 2.05-2.26 (m, 2H), 2.76-2.87 (m, 1H),
3.27 (d, 1H, J ) 5.04 Hz, OH), 4.12 (q, 2H, J ) 7.04 Hz), 4.77
(dd, 1H, J ) 5.01 and 6.70 Hz), 5.00-5.05 (m, 2H), 5.60-5.70
(m, 1H), 7.27-7.35 (m, 5H); 13C NMR (CDCl3) δ 14.69, 34.26,
53.26, 61.14, 75.26, 117.70, 126.96, 128.45, 129.09, 134.92,
142.32, 175.04; EI-MS m/z 234 (M+), 193, 143, 91, 79 (100),
77, 41. Anal. Calcd for C14H18O3: C, 71.77; H, 7.74. Found: C,
71.76; H, 7.70.
(CDCl3) δ 1.37 (s, 9H), 2.10-2.36 (m, 2H), 2.66-2.74 (m, 1H),
3.40 (bs, 1H, OH), 4.74 (t, 1H, J ) 6.17 Hz), 4.96-5.10 (m,
2H), 5.61-5.78 (m, 1H), 7.25-7.33 (m, 5H); 13C NMR (CDCl3)
δ 28.50, 38.49, 53.49, 75.14, 81.86, 117.58, 126.89, 128.22,
128.82, 135.13, 142.55, 174.45; EI-MZ m/z 206, 143, 107(100),
105, 77, 41. Anal. Calcd for C16H22O3: C, 73.25; H, 8.45.
Found: C, 73.21, H, 8.44.
2,3,3-Tr im eth yl-1-p h en ylbu ta n -1-ol (5). The reaction of
4 with BH3‚py/TiCl4 or LiEt3BH/CeCl3 gave a 98:2 or 95:5
mixture of anti-5 and syn-5, respectively. Diastereomeric
Eth yl (2R*,3S*)-3-h yd r oxy-3-p h en yl-2-p r op a r gylp r o-
purity was determined by NMR analysis: IR (neat) 3400 cm-1
;
p a n oa te (a n ti-2e): IR (neat) 3462, 1728 cm-1
;
1H NMR
EI-MS m/z 192 (M+), 118, 107 (100), 91, 57.
1
(CDCl3) δ 1.20 (t, 3H, J ) 7.02 Hz), 2.23-2.38 (m, 3H), 2.84-
2.91 (m, 1H), 3.25 (bs, 1H, OH), 4.16 (q, 2H, J ) 7.02 Hz),
4.80 (d, 1H J ) 7.63 Hz), 7.27-7.38 (m, 5H); 13C NMR (CDCl3)
δ 14.13, 18.97, 51.39, 61.17, 70.80, 73.71, 80.25, 126.36, 128.18,
128.42, 140.94, 173.48; EI-MS m/z 232 (M+), 204, 159, 115
(100), 91. Anal. Calcd for C14H16O3: C, 72.39; H, 6.94. Found:
C, 72.34; H, 6.93.
a n ti-5: H NMR (CDCl3) δ 0.62 (d, 3H, J ) 7.16 Hz), 1.02
(s, 9H), 1.75 (quint, 1H, J ) 7.12 Hz), 1.89 (bs, 1H, OH), 4.62
(d, 1H, J ) 7.43 Hz), 7.29-7.35 (m, 5H); 13C NMR (CDCl3) δ
13.36, 29.23, 23.74, 49.30, 77.70; 127.82, 127.93, 128.73,
145.46.
syn -5: 1H NMR (CDCl3) δ 0.51 (d, 3H, J ) 7.08 Hz), 1.08 (s,
9H), 1.76 (quint, 1H, J ) 7.00 Hz), 1.94 (bs, 1H, OH), 4.13 (d,
1H, J ) 6.86 Hz), 7.29-7.35 (m, 5H); 13C NMR (CDCl3) δ 11.05,
28.99, 33. 07, 50.98, 77.67, 126.91, 127.85, 128.51, 145.69.
Eth yl 2-[Hyd r oxy(p h en yl)m eth yl]-4-(2-m eth yl-1,3-d i-
oxola n -2-yl)bu ta n oa te (2g). Two diastereomeric â-hydroxy
esters were obtained in a anti/syn ratio of 64:36 (90% yield)
by the general procedure for reductions outlined above. These
diastereomers were separable by flash column chromatography
eluting with 20% EtOAc in hexanes mixture.
Ack n ow led gm en t. Work carried out in the frame-
work of the National Project “Stereoselezione in Sintesi
Organica. Metodologie e Applicazioni” was supported by
the Ministry of University and Technological Research
(MURST) Rome and by the University of Camerino. The
authors thank Dr. Gianni Rafaiani for his help in the
NMR experiments, also.
1
An ti d ia ster eom er : IR (neat) 3458, 1725 cm-1; H NMR
(CDCl3) δ 1.27 (t, 3H, J ) 6.95 Hz), 1.32 (s, 3H), 1.45-1.62
(m, 2H), 1.83 (t, 2H, J ) 7.68 Hz), 3.12-3.17 (m, 1H), 3.80-
3.90 (m, 4H), 3.98 (bs, 1H, OH), 4.14 (q, 2H, J ) 6.95 Hz),
4.75 (d, 1H, J ) 7.40 Hz), 7.12-7.38 (m, 5H); 13C NMR (CDCl3)
δ 14.19, 23.56, 24.04, 36.97, 52.29, 60.51, 64.28, 64.84, 72.77,
J O9821574