Studies on diastereoselective reduction of cyclic β-ketoesters with boron hydrides. Part 4: The reductive profile of functionalized cyclohexanone derivatives

Article abstract of DOI:10.1016/j.tet.2004.01.079

Reduction of 2-allyl-2-carboalkoxycyclohexanones (3d-f), 2-propyl-2-carboethoxycyclohexanone (3g) and 2-benzyl-2- carboethoxycyclohexanone (3h) with boron hydrides in the presence and absence of several chelating agents were studied. Molecular modeling studies using semiempirical PM3 method were performed in order to find a suitable explanation of the diastereoselection of ketone carbonyl faces during the reductive process, which yielded trans-2-allyl-2-carboethoxycyclohexanol (6e) and cis-2-allyl-2-carboethoxycyclohexanol (7e) in good diastereomeric excess by using inexpensive sodium and tetrabutylammonium borohydrides.

Full text of DOI:10.1016/j.tet.2004.01.079

Tetrahedron 60 (2004) 2745–2755
Studies on diastereoselective reduction of cyclic b-ketoesters with
boron hydrides. Part 4: The reductive profile of functionalized
cyclohexanone derivatives
a,b
Carlos A. M. Fraga,a,b Lis Helena P. Teixeira, Carla Maria de S. Menezes,a
Carlos Mauricio R. Sant’Anna,a,c Maria da Conceic¸a˜o K. V. Ramos,b
Francisco R. de Aquino Netob and Eliezer J. Barreiroa,b
,
*
a
´ ´ ˆ ´
Laboratorio de Avaliac¸a˜o e Sıntese de Substancias Bioativas (LASSBio), Faculdade de Farmacia, Universidade Federal do Rio de Janeiro,
PO Box 68023, 21944971 Rio de Janeiro, RJ, Brazil
Instituto de Quımica, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil
b
´
Departamento de Quımica, ICE, Universidade Federal Rural do Rio de Janeiro, RJ 23851-970 Seropedica, Brazil
c
´
´
Received 3 December 2003; revised 14 January 2004; accepted 26 January 2004
Abstract—Reduction of 2-allyl-2-carboalkoxycyclohexanones (3d-f), 2-propyl-2-carboethoxycyclohexanone (3g) and 2-benzyl-2-
carboethoxycyclohexanone (3h) with boron hydrides in the presence and absence of several chelating agents were studied. Molecular
modeling studies using semiempirical PM3 method were performed in order to find a suitable explanation of the diastereoselection of ketone
carbonyl faces during the reductive process, which yielded trans-2-allyl-2-carboethoxycyclohexanol (6e) and cis-2-allyl-2-carboethoxy-
cyclohexanol (7e) in good diastereomeric excess by using inexpensive sodium and tetrabutylammonium borohydrides.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Cyclic b-ketoesters and its corresponding b-hydroxyester
derivatives are important building blocks for the synthesis
of natural products and many bioactive substances.1,2
Several previous papers of our laboratory described the
enantio-3,4 and diastereoselective5–8 preparation of cyclic
b-hydroxyester derivatives, exploring the chemoselective
reduction of the carbonyl group of 2-alkyl-2-carboalkoxy-
cyclopentanone (1) and 2-acetyl-2-alkyl-butyrolactone
derivatives (2) and their application in the synthesis of
carbocyclic and heterocyclic building blocks9–12 useful for
construction of new drug candidates13,14 (Fig. 1).
Now, we describe in this paper our studies about the
reduction of 2-alkyl-2-carboalkoxycyclohexanone deriva-
tives (3d-m) using boron hydrides and the investigation of
Figure 1. Cyclic b-ketoester derivatives.
its diastereoselection profile in comparison with those
present by corresponding cyclopentanone derivatives (1).
Additionally, molecular modeling studies were performed
in derivatives of both series (1) and (3) in order to elucidate
the structural reasons of their reductive profile (Fig. 1).
Keywords: Diastereoselective reduction; Boron hydrides; Ketoesters.
*
´
´
Corresponding author. Address: Laboratorio de Avaliac¸a˜o e Sıntese de
2. Results and discussion
ˆ
´
Substancias Bioativas (LASSBio), Faculdade de Farmacia, Universidade
Federal do Rio de Janeiro, PO Box 68023, 21944971 Rio de Janeiro, RJ,
Brazil. Tel./fax: þ55-21-25626644;
The 2-carboalkoxcyclohexanone derivatives (3a), (3b) and
(3c) were synthesized in 90, 71 and 95% yield, respectively,
e-mail address: cmfraga@pharma.ufrj.br
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.01.079

2746
C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
Scheme 1. (i) ROH, H2SO4 (cat.), reflux, 4–8 h, 98% (5a), 82% (5c); (ii) AlCl3, Et3N, rt, 2.5–4 h, 90% (3a), 71% (3b), 95% (3c).
exploiting the Dieckmann condensation of the correspond-
ing pimelic esters (5a-c) (Scheme 1) by treatment with
AlCl3 and triethylamine.15 In spite of ethyl pimelate (5b)
having been obtained commercially,16 the corresponding
methyl and isobutyl esters were obtained in 98 and 82%
yield respectively from Fisher esterification of pimelic acid
(4).17
HRGC using a b-cyclodextrin derivative as stationary
phase,20,21 as illustrated in Figure 2. The chiral HRGC
method was elected instead of normal phase GC due to the
better resolution profile previously evidenced for the dia-
stereomeric separation of functionalized b-ketoesters.20,21
In spite of Frater22 having described that the reduction of
2-allyl-2-carboethoxycyclohexanone (3e) with sodium
borohydride in ethanol furnished the cis-cyclohexanol
derivative (7e) in 33% de, we found that the use of
methanol as solvent (entry 3, Table 1) produces a slight
improvement of this diastereoselectivity profile, resulting
also in the major formation of (7e), but in 43% de. The
previous addition of 2 equiv. of CaCl2 (entry 4, Table 1)
resulted in the inversion of the relative configuration of the
major isomer produced, that is, cis-cyclohexanol derivative
(6e).
When we employed reactive halides like allyl bromide and
benzyl bromide, C-2 alkylated derivatives (3d-f) and (3h)
were regioselectively obtained in yields ranging from 53 to
88%, by using a modification18 of the classical Barco’s
conditions19 which avoids the formation of the correspond-
ing O-alkylated derivatives. Additionally, the alkylation of
2-carboethoxycyclohexanone (3b) with less reactive propyl
bromide furnished the desired C-alkylated derivative (3g) in
only 24% yield (Scheme 2).
Next, the 2-allyl-2-carboalkoxycyclohexanone derivatives
(3d-f) were submitted to the chemoselective reduction with
sodium borohydride in methanol, in the presence or absence
of calcium chloride, in order to compare its reductive profile
with that previously described for the corresponding 2-allyl-
2-carboalkoxycyclopentanone derivatives5 (1d-f), as
showed in Table 1. The composition of the diastereomeric
alcohols mixture (6 and 7) was elucidated by NMR
Next, we investigated the contribution of the size of ester-
attached alkoxy group in the diastereoselective reductive
profile of the 2-allyl-2-carboalkoxy-cyclohexanone deriva-
tives (3d-f) with sodium borohydride (Table 1), following
an experimental evidence related in a previous paper
from our laboratory,5 which indicated that the diastereo-
selectivity of the 2-allyl-2-carboalkoxycyclopentanone
(1d-f) reduction was inversely proportional to the bulky
of the alkoxy group. In fact, in that work the best
diastereoselectivity index was achieved in the reduction of
5,7
spectroscopy using growing concentrations of Eu(thd)3
and the relative diastereomeric ratio was determined by
Scheme 2.

C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
2747
Table 1. Reduction of (^)-2-allyl-2-carboalkoxycyclohexanone derivatives (3d-f) with sodium borohydride
Entry
Compound
R
Conditionsa
Product 6:7
Yield (%)
Diastereomeric ratiob,c trans/cis
1
2
3
4
5
6
3d
3d
3e
3e
3f
Me
Me
Et
Et
iBu
iBu
A
B
A
B
A
B
6d:7d
6d:7d
6e:7e
6e:7e
6f:7f
74
85
90
96
92
96
1:1.6
4.3:1
1:2.5
6.5:1
1:2.4
2.2:1
3f
6f:7f
a
Conditions: (A) NaBH4 (1.2 equiv.), MeOH, 0 8C, 30 min; (B) (i) CaCl2 (2 equiv.), MeOH, rt, (ii) NaBH4 (1.2 equiv.), 0 8C, 30 min.
The relative diastereomeric ratio was determined by HRGC in a 10% 2,3-di-O-methyl-6-O-t-butyldimethylsilyl-b-cyclodextrin in SE-54 capillary column
(20 m£0.3 mm£0.3 mm).
b
c
1
The qualitative determination of diastereomeric alcohols was made by analysis of H NMR at 200 MHz, in presence of Eu(thd)3.
of the respective complexes with the calcium ion. The
optimized structures of (3d) and of its calcium complex
(3d)þCa2þ are presented in Figure 4. The comparison of
these structures, especially the van der Waals radius
representation, indicates a change in the conformation of
the keto, alkoxy ester, and allyl groups in (3d) that allows
the molecule to act as a tridentate ligand, in order to stabilize
the positive charge of the calcium ion. This geometry is in
Figure 2. Chiral-HRGC of stereoisomers of the 2-allyl-2-carbomethoxy-
cyclohexanol derivatives (6d) and (7d), at 100 8C, in a capillary glass
column (20 m£0.3 mm£0.3 mm) covered with 10% of 2,3-dimetil-6-O-
terc-butildimetilsilil-b-cyclodextrina/SE-54. A¼(^)-cis-2-allyl-2-carbo-
methoxycyclohexanols (7d); B¼(^)-trans-2-allyl-2-carbomethoxycyclo-
hexanols (6d).
accordance with the calculated geometry previously
reported for the complex of (1d) with a Zn2þ ion.6
Electronic surfaces as MEP and frontier orbitals maps are
useful theoretical tools to evaluate the 3D electronic
properties of a compound, and with its associated molecular
size and shape, may be of great value to interpret, elucidate
and predict experimental results of stereoselective organic
reactions.24,25 The MEP illustrates the most (red) and less
(blue) electron-rich regions, throughout an energetic
gradient illustrated by a red–orange–yellow–green–blue
scale. Also considering a similar color scale, the LUMO
(Lowest Unoccupied Molecular Orbital) map shows the
absolute value of the LUMO onto the total electron density
surface.
2-allyl-2-carbomethoxycyclopentanone (1d) with sodium
borohydride in the absence or in the presence of calcium
chloride5,6 (Fig. 2). However, in the present work the
diastereoselection of the ketone carbonyl faces of 2-allyl-2-
carboalkoxycyclohexanone derivatives (3d-f) by hydride
anion have no relationship with the size of the alkoxy group
(Table 1) since the diastereomeric excess followed the order
(3e).(3d).(3f). As showed previously, the best dia-
stereoselective control was achieved during the reduction
of the ethyl ester (3e) with NaBH4/CaCl2, which produced
the cis-(^)-2-allyl-2-carboethoxycyclohexanol (7e) in 73%
de. Intriguingly, despite applying the same experimental
conditions, the diastereocourse of the reduction of the
cyclohexanone derivatives with sodium borohydride in the
absence or in the presence of CaCl2 was opposite to that of
the corresponding cyclopentanone derivatives (1d-f), as
illustrated in Figure 3 for the reduction of methyl ester (1d).
The MEPs observed onto the Re and Si faces of (3d)
(Fig. 5(A)) allow the recognition of an electron-poorer
region (a more intense blue color) near the Re face of the
ketone carbonyl group (see Fig. 5(D)) than on the
corresponding region on the Si face (a green colored
region). The LUMO maps (Fig. 5(B)) also indicate the Re
face of the ketone carbonyl (a much more intense blue
color) as the most favorable to suffer the attack of a
nucleophile26,27 (hydride anion). In addition, an electron-
rich region can be seen on the Si face (a yellow–orange
colored region in the MEP and a red–orange colored region
in the LUMO map), corresponding to the oxygen atom of
the alcoxy ester group, which may render unfavorable the
approximation of a nucleophile due to electrostatic repul-
sion. Moreover, the orientation of the alkoxy ester group
may also interfere with the borohydride approximation on
the Si face by steric hindrance over the ketone carbonyl. In
In order to elucidate the possible reasons for the inversion of
the diastereoselective reductive profile as consequence of
the ring homologation, we made a comparison of the
structural and electronic properties of the substrate
molecules using the semiempirical PM3 method.23 All
optimized PM3 structures obtained in the group of 2-allyl-2-
carboalkoxycyclohexanones (3d-f) had similar geometries,
independently of the size of the ester-attached alkoxy group
(OR). Similar geometries were also observed for each group

2748
C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
Figure 3. Inversion of the diastereoselective profile of the cyclic b-hydroxyesters (6-8) obtained from reduction of 2-allyl-2-carbomethoxycyclopentanone
derivative (1d) or 2-allyl-2-carboalkoxycyclohexanone derivatives (3d-f) with sodium borohydride in the presence or absence of calcium chloride.
contrast, the electronic effect of the allyl group on the Re
face should not greatly influence the nucleophilic attack
because it results in a less intense electron-rich region than
the alkoxy ester group and it is oriented in the opposite
direction in relation to the ketone carbonyl group.
The MEP and LUMO maps of the compounds complexed
with the calcium ion, exemplified in the Figure 5 by
(3d)þCa2þ, show the region corresponding to the Si face on
the ketone carbonyl group (Fig. 5(A.1) and (B.1)) as a
relatively electron-poor region (orange–green color). On
the other hand, the Re face of the ketone group is now
almost completely hindered by the allyl group because of its
conformational change towards the Ca2þ ion, which is
better illustrated by the space-filling model representation
(Fig. 5(C.1)). This steric hindrance would probably have a
stronger effect on the nucleophilic attack than the electro-
static repulsion created by the oxygen atoms of the alkoxy
ester group (red and red–orange regions on the MEP and
LUMO maps, respectively) observed on the Si face. As a
Figure 4. Tube and space-filling representations showing the confor-
result, in opposition to uncomplexed (3d), the Re face of its
mational changes in the keto, alcoxy ester, and allyl groups of (3d) that
complex with Ca2þ is much less susceptible to approxi-
allows it to interact with the calcium ion as a tridentate ligand in the
(3dþCa2þ) complex. Color code for the atoms is: carbon, gray; oxygen,
mation and, consequently, to nucleophilic attack. Therefore,
the (^)-trans-cyclohexanol derivatives (6) must be obtained
red; and calcium, yellow. The hydrogen atoms were omitted for clarity.

C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
2749
Figure 5. Representations of (3d) and (3d)þCa2þ. (A) MEPs are in the range of 254 (red) to þ20 (blue) kcal/mol; (B) LUMO maps in the range of 1.1027
(red) to 0.030 (blue) kcal/mol; (C) space-filling model representation; (D) tube model representation; (A.1) MEPs in the range of þ115 (red) to þ380 (blue)
kcal/mol; (B.1) LUMO maps in the range of 7.1028 (red) to 0.0090 (blue) kcal/mol; (C.1) space-filling representation; (D.1) tube model representation. Color
code for the atoms is: carbon, gray; oxygen, red; and calcium, yellow. The hydrogen atoms were omitted for clarity.
in greater proportion than the corresponding (^)-cis-
cyclohexanol derivatives (7), which is in agreement with
the experimental results (Table 1).
reduction of the respective allyl derivatives (6d) and (7d)
(Table 1), that is, 26% (entry 7, Table 2) versus 42% (entry
3, Table 1) without the use of CaCl2 and 62% de (entry 8,
Table 2) versus 73% (entry 4, Table 1) when CaCl2 was
used as a complexing agent (entry 8). These results
indicated to us that the change of the allyl to propyl group
in the derivative (3g) influenced directly of the dia-
stereocourse of this reaction, by adopting a particular
conformation that could partially block the less hindered Re
face of the keto-carbonyl group or by the absence of
formation of a ternary complex with calcium ion, as
anticipated by molecular modeling studies.
In order to evaluate the effect of the allyl group in the
diastereoselectivity of the ethyl ester derivative (3e)
(Table 1, entries 3 and 4) we studied the reductive profile
of the corresponding saturated propyl derivative (3g) and
the benzyl analogue (3h) with sodium borohydride in the
presence or absence of calcium chloride. As depicted in
Table 2, the reduction of saturated derivative (3g) resulted
in the formation of a mixture of the cyclohexanol
derivatives (6g) and (7g) with a decrease of the dia-
stereomeric excess in comparison to that obtained from the
The reduction of benzyl derivative (3h) with sodium
Table 2. Reduction of (^)-2-propyl-2-carboethoxycyclohexanone (3g) and (^)-2-benzyl-2-carboethoxycyclohexanone derivatives (3h) with sodium
borohydride
Entry
Compound
R1
Conditionsa
Product 6:7
Yield (%)
Diastereomeric ratiob,c trans/cis
7
8
9
10
3g
3g
3h
3h
nPr
nPr
Bn
Bn
A
B
A
B
6g:7g
6g:7g
6h:7h
6h:7h
91
98
98
90
1:1.7
4.3:1
1:2.4
2.9:1
a
Conditions: (A) NaBH4 (1.2 equiv.), MeOH, 0 8C, 30 min; (B) (i) CaCl2 (2 equiv.), MeOH, rt, (ii) NaBH4 (1.2 equiv.), 0 8C, 30 min.
The relative diastereomeric ratio was determined by HRGC in a 10% 2,3-di-O-methyl-6-O-t-butyldimethylsilyl-b-cyclodextrin in SE-54 capillary column
(20 m£0.3 mm£0.3 mm).
b
c
1
The qualitative determination of diastereomeric alcohols was made by analysis of H NMR at 200 MHz, in presence of Eu(thd)3.

2750
C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
borohydride without CaCl2 (Table 2, entry 9) did not result
in any change of the diastereoselectivity in comparison with
that showed by the corresponding allyl derivative (3e)
(Table 1, entry 3). On the other hand, the use of calcium
chloride as Lewis acid (Table 2, entry 10) resulted in a
expressive drop of the comparative diastereoselection (see
Table 1, entry 4), indicating that possibly for steric reasons
benzyl group is not so able to adopt the adequate orientation
which permits the formation of the ternary complex between
the allyl group of compound (3e) and calcium ion, reducing
selective blockage on the Si face of the ketone carbonyl
group.
the reductions of the allyl derivative (3e) using these other
metallic chlorides have not been able to improve the
diastereoselectivity, the diastereofacial discrimination of
the nucleophilic hydride attack to the ketone group was
abolished, resulting in the equal formation of the isomers
trans-(6e) and cis-(7e) (Table 3, entries 14–17). The
evidenced profile indicated that, contrarily to the results
previously described by Taniguchi et al.,29 the reduction of
derivative (3e) in the presence of Lewis acids did not show a
direct correlation between the ionic radius of the metal and
the diastereoselectivity of the reductive process.
Finally, the last variation in the reductant conditions
consisted in the employment of other boron hydrides with
different reactivity, size and solubility in the aprotic solvent
THF, represented by the use of zinc borohydride,30 lithium
tri-sec-butylborohydride31 (L-Selectride) and tetrabutyl-
ammonium borohydride32 (Table 3, entries 18–20, respect-
ively). In spite of being well-known that the use of bulky
boron hydrides led to an improvement of the diastereofacial
discrimination of carbonyl ketone group, the treatment of
allyl functionalized derivative (3e) either with zinc
borohydride or L-Selectride in THF resulted in an almost
complete absence of the diastereoselectivity between the
cyclohexanol derivatives (6e) and (7e) (Table 3, entries 18
and 19), indicating to us that bulky hydride-containing
species are not able to discriminate the faces of the ketone
carbonyl group.
Considering the results described herein, we elected the
2-allyl-2-carboethoxy-cyclohexanone derivative (3e) for
further studies varying the conditions of the reductive step
as well as the nature and the size of the hydride transferring
reagent, in order to optimize the diastereoselective for-
mation of cis-2-allyl-2-carboethoxycyclohexanol (7e),
obtained only in very poor de.
The initial modification, which consisted in the change of
methanol used as solvent to isopropanol (Table 3, entries 11
and 12) or aprotic tetrahydrofuran (Table 3, entry 3), led to
the loss of the diastereoselectivity evidenced before. This
distinct profile may be explained by the presence of different
reducing species in the media, since sodium borohydride
reacts with methanol to give sterically demanding tri-
methoxyborohydride, whereas solutions of sodium boro-
hydride in isopropanol or tetrahydrofuran are very stable.28
On the other hand, the reduction of derivative (3e) with
tetrabutylammonium borohydride in THF, furnished the cis-
cyclohexanol derivative (7e) with the desired improvement
of the diastereoselectivity from 42% (Table 1, entry 3) to
68% de (Table 3, entry 20). The preferential formation of
the diastereomer (7e) by the usage of a (nBu)4NBH4 in THF
(Table 3, entry 20) can be explained by the better solubility
of the non-bulky hydride species in the aprotic media, which
Once that we obtained the best diastereoselective excess by
using CaCl2 as complexing agent, we decide next
investigate the profile of the reductive process with sodium
borohydride employing four other Lewis acids presenting a
variation of the atomic radius of the divalent metal from
Mgþ2 to Znþ2 (Table 3, entries 14–17). However, besides
Table 3. Reduction of 2-allyl-2-carboethoxycyclohexanone derivative (3e)
Entry
Redutor
Lewis acid
Solvent and temperature (8C)
Product 6:7
Yield (%)
Diastereomeric ratioa,b trans/cis
11
12
13
14
15
16
17
18
19
20
21
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
NaBH4
Zn(BH4)2
L-Selectride
(nBu)4NBH4
(nBu)4NBH4
—
CaCl2
—
MgCl2
MnCl2
CeCl3
ZnCl2
—
iPrOH, 0 8C
iPrOH, 0 8C
THF, 0 8C
MeOH, 0 8C
MeOH, 0 8C
MeOH, 0 8C
MeOH, 0 8C
THF, 0 8C
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
6e:7e
85
90
77
98
96
91
98
91
95
86
89
1:1.2
1:1.1
1:1
1.6:1
1:1.3
1.2:1
1:1
1.2:1
1:1.1
1:5.3
1:8.2
—
—
—
THF, 278 8C
THF, 0 8C
MeOH, 0 8C
a
The relative diastereomeric ratio was determined by HRGC in a 10% 2,3-di-O-methyl-6-O-t-butyldimethylsilyl-b-cyclodextrin in SE-54 capillary column
(20 m£0.3 mm£0.3 mm).
b
1
The qualitative determination of diastereomeric alcohols was made by analysis of H NMR at 200 MHz, in presence of Eu(thd)3.

C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
2751
raised the speed of the reaction favoring the attack on the
less hindered Si face of ketone carbonyl group (Fig. 5).
Nevertheless, the change of the solvent from aprotic THF to
protic methanol (Table 3, entry 21) curiously increased the
diastereoselective formation of alcohol (7e) to 78% de. In
fact, in spite of there are not many works in the literature
describing the use of (nBu)4NBH4 in protic to solvents for
the reduction of carbonyl compounds29,33 is well-known
that its application in the reduction of b-ketoesters29
followed the Felkin–Ahn’s model34 with the attack of the
hydride anion at the less hindered face of ketone carbonyl
group (Fig. 6), in agreement with the molecular modeling
and chemical results obtained in the present work.
Monte Carlo conformational analysis with the PM3 method
was employed. The minimal energy conformers were
selected and submitted to single-point energy calculations
with the ab initio 3-21Gp basis set in order to better evaluate
their electronic properties. In this study, the map of the
electrostatic potential (MEP), and the map of the absolute
value of the lowest-unoccupied molecular orbital (LUMO
map), both onto an electron density surface of 0.002 e/au3,
were considered for the analysis of the stereoselectivity
results obtained in the Section 4.
4.2. Chemistry
1H and 13C NMR spectra were determined in deuterochloro-
form containing ca. 1% tetramethylsilane as an internal
standard with Brucker AC 200 and Varian VxR 300
spectrometers. Splitting patterns were as follows: s, singlet;
d, doublet; t, triplet; q, quartet; qt, quintet; dd, double
doublet; ddt, double double triplet; m, multiplet. Infrared
spectra (IR) spectra were obtained with a Nicolet 505
Magna spectrophotometer as neat films on sodium chloride
plates. The mass spectra (MS) were obtained on a GC/VG
Micromass 12 at 70 eV. Gas chromatography (HRGC) was
recorded in a Hewlett Packard model 5890 series II using
injection in the split mode. The HRGC analyses were
performed in 10% 2,3-di-O-methyl-6-O-t-butyldimethyl-
silyl-b-cyclodextrin in SE-54 (1% vinyl; 5% phenyl; 94%
methylpolysiloxane) house made capillary column
(20 m£0.3 mm£0.3 mm) at 100 8C/2 8C/min/130 8C.
Microanalysis data were obtained with a Perkin–Elmer
240 analyzer, using Perkin–Elmer AD-4 balance.
Figure 6. Felkin–Ahn’s attack of hydride anion to less hindered face of
cyclohexanone derivative (3e).
3. Conclusion
The progress of all reactions was monitored by tlc which
was performed on 2.0 cm£6.0 cm aluminum sheets pre-
coated with silica gel 60 (HF-254, Merck) to a thickness of
0.25 mm. The developed chromatograms were visualized
with molybdatophosphoric acid in ethanol. For column
chromatography Merck silica gel (70–230 mesh) was used.
Solvents used in the reactions were redistilled prior use and
In summary, the results obtained from this research work
furnished a nice approach to the diastereoselective synthesis
of (^)-trans-cyclohexanol derivative (6e) and (^)-cis-
cyclohexanol derivative (7e) respectively in 73 and 78%
de, using available and inexpensive sodium or tetrabutyl-
ammonium borohydrides. The developed synthetic method-
ologies showed to be extremely dependent of the solvent
and the Lewis acid employed, being the best results obtained
when the reductions were carried out in methanol and, for
the preparation of (6e), calcium chloride was used as Lewis
acid.
˚
stored over 3–4 A molecular sieves. Reactions were
generally carried out under nitrogen atmosphere and
magnetic stirring. The ‘usual workup’ means that the
organic extracts prior to concentration, under reduced
pressure, were treated with a saturated aqueous sodium
chloride solution, dried over anhydrous sodium sulfate and
filtered.
4. Experimental
4.3. General procedure for the esterification of pimelic
acid (4)
4.1. Molecular modeling
The molecular modeling analysis was performed using the
SPARTAN 1.0.5 program (Wavefunction Inc., Irvine, CA,
2000) on a Pentium III 900 MHz computer. The structure of
the compounds (3d-f) and (1d) and of their respective
To a solution of pimelic acid15 (4) (1 g; 6.25 mmol) in
12 mL of methyl or isobutyl alcohol was slowly added
0.73 mL of concentrated sulfuric acid. The resulting mixture
was stirred at reflux until that tlc analysis indicated the total
consumption of the starting material (eluent: hexanes/
AcOEt 70:30). Next, the mixture was poured into crushed
ice and then, extracted with dichloromethane (5£50 mL).
The organic layers were washed with 5% aq. NaHCO3
solution and submitted to the ‘usual workup’ to furnish the
corresponding pimelate ester (5a) or (5c) as described
above.
complexes with calcium ion, (3d-f)þCa2þ and (1d)þCa2þ
,
were optimized with the PM3 method.23 This semiempirical
method is parameterized for calcium, present in the Lewis
acid CaCl2 used in the experimental methodology, and was
previously used to analyze diastereoselective experimental
data.12,35
The optimized structures of the compounds were submitted
to Hessian matrix analysis to unequivocally characterize
them as true minima of the potential energy surface. A
4.3.1. Dimethyl pimelate (5a). The spectroscopic data

2752
C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
of this compound, which was obtained in 99% yield,
are in agreement with those previously related in
literature.36
stirred at room temperature until that tlc analyses (Hex/
AcOEt, 9:1) indicated the total consumption of the starting
material. The suspension was filtered, the filtrate concen-
trated at reduced pressure (80 mm Hg) and the residue
diluted with ether (50 mL). The ‘usual workup’ gives the
respective 2-alkyl-2-carboalkoxycyclohexanone derivative
(3d-g).
4.3.2. Diisobutyl pimelate (5c). This compound was
obtained in 82% yield, after 8 h, as a yellow oil; IR (film):
n C–H 2875 and 2961, n CvO 1737, n C–O 1175 cm21
;
1H NMR (200 MHz): 0.92 (d, 12H, J¼6.7 Hz, OCH2-
CH(CH3)2), 1.38 (m, 2H, OvCCH2CH2CH2CH2CH2CvO),
1.64 (qt, 4H, J¼7.5 Hz, OvCCH2CH2CH2CH2CH2CvO),
1.89 (sp, 2H, J¼6.7 Hz, OCH2CH(CH3)2), 2.31 (t, 4H,
J¼7.5 Hz, OvCCH2CH2CH2CH2CH2CvO), 3.85 (d, 4H,
J¼6.7 Hz, OCH2CH(CH3)2) ppm; 13C NMR (75 MHz):
19.2 (OCH2CH(CH3)2), 24.8 (CH2-4), 27.8 (OCH2-
CH(CH3)2), 28.8 (CH2-3 and CH2-5), 34.3 (CH2-2 and
CH2-6), 70.6 (OCH2CH(CH3)2), 173.8 (CvO) ppm.
Anal. calcd for C15H28O4: C 66.14; H 10.36. Found: C
66.21; H 10.44.
4.5.1. 2-Allyl-2-carbomethoxycyclohexanone (3d). The
spectroscopic data of this compound, which was obtained in
53% yield, are in agreement with that previously related in
literature.40
4.5.2. 2-Allyl-2-carboethoxycyclohexanone (3e). The
spectroscopic data of this compound, which was obtained
in 53% yield, are in agreement with that previously related
in literature.22,40
4.5.3. 2-Allyl-2-carboisobutoxycyclohexanone (3f). From
alkylation of (3c) with allyl bromide (0.66 mL), this
compound was obtained in 88% yield,41 after 24 h, as a
yellow oil; IR (film): n CvC–H 3078, n C–H 2960 and
4.4. General procedure for Dieckmann cyclization16 of
the esters (5a-c)
To a suspension of anhydrous aluminum chloride (16 g,
120 mmol) in 50 mL of dichloromethane was added a
solution of the corresponding pimelate ester derivative
(5a-c) (46 mmol) in 50 mL dichloromethane. After cooling
the obtained mixture at 0 8C, 16 mL of triethylamine
(120 mmol) was carefully added and reaction was stirred
at room temperature until that tlc analysis indicated the total
consumption of the starting material. Next, a 1:1 mixture of
10% aq. HCl and crushed ice (100 mL) was added and the
reaction was extracted with dichloromethane (4£40 mL).
The organic layers were washed with a saturated aq. oxalic
acid solution and submitted to the usual workup to furnish
the corresponding 2-carboalkoxycyclohexanone derivative
(3a-c) as described next.
2873, n CvO 1737 and 1716, n C–O 1219 and 1203 cm21
;
1H NMR (200 MHz): 0.86 (d, 6H, J¼6.7 Hz, COOCH2-
CH(CH3)2), 1.30–2.00 (m, 6H, OvCCH2CH2CH2CH2;
COOCH2CH(CH3)2 and CHHCHvCH2), 2.33–2.51 (m,
5H, OvCCH2CH2CH2CH2 and CHHCHvCH2), 3.83 (d,
2H, J¼6.6 Hz, 1H, COOCH2CH(CH3)2), 4.93–5.01 (m, 2H,
CH2CHvCH2), 5.62–5.76 (m, 1H, CH2CHvCH2) ppm;
13C NMR (50 MHz): 207.3 (CvO), 171.5 (COOCH2-
CH(CH3)2), 133.4 (CH2CHvCH2), 118.1 (CH2CHvCH2),
71.3 (COOCH2CH(CH3)2), 61.0 (C-2), 41.1 (CH2-
CHvCH2), 39.4 (CH2-6), 35.7 (CH2-3), 27.7 (COOCH2-
CH(CH3)2), 27.5 (CH2-5), 22.4 (CH2-4), 19.1
(COOCH2CH(CH3)2) ppm. Anal. calcd for C14H22O3: C
70.56; H 9.30. Found: C 70.67; H 9.35.
4.4.1. 2-Carbomethoxycyclohexanone (3a). The spectro-
scopic data of this compound, which was obtained in 90%
yield, are in agreement with those previously related in
literature.37
4.5.4. 2-Benzyl-2-carboethoxy-cyclohexanone (3h). From
alkylation of (3b) with benzyl bromide (0.83 mL), this
compound was obtained in 80% yield, after 12 h, as a
yellow oil; IR (film): n –CvC–H 3085, 3062 and 3029, n
C–H 2942 and 2867, n CvO 1740 and 1714, n C–O
1
4.4.2. 2-Carboethoxycyclohexanone (3b). The spectro-
scopic data of this compound, which was obtained in 71%
yield, are in agreement with that previously related in
literature.38,39
1188 cm21; H NMR (200 MHz): 1.16 (t, 3H, J¼7.1 Hz,
COOCH2CH3), 1.65–1.71 (m, 4H OvCCH2CH2CH2CH2),
1.80–2.00 (m, 3H, OvCCH2CH2CHCHH-syn to benzyl
group), 2.37–2.47 (m, 3H, OvCCH2CH2CH2CHH-anti to
benzyl group), 2.86 (d, 2H, J¼13.7 Hz, PhCHH), 3.30 (d,
2H, J¼13.7 Hz, PhCHH), 4.08 (q, 2H, J¼7.1 Hz, COOCH2-
CH3), 7.08–7.12 (m, 2H, metaAr-H), 7.18–7.26 (m, 3H,
orthoAr-H and paraAr-H) ppm; 13C NMR (50 MHz): 207.3
(CvO), 171.1 (COOCH2CH3), 136.7 (ipsoAr), 130.4
(orthoAr), 128.0 (metaAr), 126.7 (paraAr), 62.2 (C-2),
61.3 (COOCH2CH3), 41.4 (–CH2Ph), 40.5 (CH2-6), 36.0
(CH2-3), 27.7 (CH2-5), 22.6 (CH2-4), 14.0 (COOCH2CH3)
ppm. Anal. calcd for C16H20O3: C 73.82; H 7.74. Found: C
73.77; H 7.81.
4.4.3. 2-Carboisobutoxycyclohexanone (3c). The spectro-
scopic data of this compound, which was obtained in 95%
yield, are in agreement with that previously related in
literature.38
4.5. General procedure for the C-alkylation of the
b-ketoesters (5a-c)17,18
To a suspension of anhydrous potassium carbonate (2.44 g;
17.6 mmol) in anhydrous acetone (6 mL) was added a
solution of 2-carboalkoxycyclohexanone derivative (5a-c)
(5.8 mmol) in anhydrous acetone (2 mL). The reaction
mixture displays a characteristic yellow color after stirring
at room temperature for 30 min due to the formation of the
corresponding enolate intermediate. Then, respective alkyl
bromide (7.6 mmol) was added slowly and the mixture was
4.5.5. 2-Propyl-2-carboethoxycyclohexanone (3g). From
alkylation of (3b) with n-propyl bromide (0.65 mL), this
compound was obtained in 24% yield, after 120 h, as a
yellow oil; IR (film): n C–H 2961, 2939 and 2372, n CvO
1
1732 and 1715, n C–O 1204 cm21; H NMR (200 MHz):
0.90 (m, 5H, CH2CH2CH3), 1.22–1.29 (m, 5H, COOCH2CH3

C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
2753
and CH2CH2CH3), 1.39–1.68 (m, 6H, OvCCH2CH2-
CH2CH2), 2.26–2.33 (m, 2H, OvCCH2CH2CH2CH2),
3.99–4.21 (m, 2H, COOCH2CH3) ppm; 13C NMR
(50 MHz): 205.3 (CvO), 173.7 (COOCH2CH3), 61.2
(C-2), 60.3 (COOCH2CH3), 41.3 (CH2-6), 34.3 (CH2-3),
27.8 (CH2-5), 24.7 (CH2CH2CH3), 22.7 (CH2-4), 17.7
(CH2CH2CH3), 14.4 (COOCH2CH3 and CH2CH2CH3)
ppm. Anal. calcd for C12H20O3: C 67.89; H 9.50. Found:
C 67.93; H 9.43.
affording the mixture of diastereomeric alcohols (6e) and
(7e) as described Table 3.
4.9. General procedure for reduction of 2-allyl-2-carbo-
ethoxy-cyclohexanone (3e) with tetrabutylammonium
borohydride
To a solution of b-ketoester derivative (3e) (1 mmol) in
solvent (methanol or anhydrous THF) (6 mL), cooled at 08,
was slowly added 0.325 g (1.2 mmol) of tetrabutyl-
ammonium borohydride. The mixture was stirred at 0 8C
(or 278 8C) for 30 min, when the solvent was concentrated
at reduced pressure (80 mm Hg). The white doughy residue
was diluted with methylene chloride (30 mL). The resulting
solution was washed with 1 N aq. HCl solution (30 mL).
Usual workup of the organic layer afforded the mixture of
diastereomeric alcohols as described in Table 3.
4.6. General procedure for reduction of 2-alkyl-2-
carboalkoxy-cyclohexanone derivatives (3d-g) with
sodium borohydride, in the presence or in the absence of
metallic halides
A solution of b-ketoester derivative (3d-g) (1 mmol) in
solvent (methanol, isopropanol or THF) (6 mL), in the
presence or absence of the anhydrous metallic halide
(2 mmol), was stirred at room temperature for 30 min.
The reaction mixture was cooled at 0 8C (or 278 8C), and
0.045 g (1.2 mmol) of sodium borohydride was slowly
added. A clear solution was obtained, which was stirred at
0 8C (or 278 8C) for 30 min. The solvent was concentrated
at reduced pressure (80 mm Hg). The white doughy residue
was diluted with methylene chloride (30 mL). The resulting
solution was washed with saturated aqueous ammonium
chloride solution (30 mL). Usual workup of the organic
layer afforded the mixture of diastereomeric alcohols as
described in Tables 1–3.
4.9.1. Diastereomeric mixture of trans- and cis-(6)-2-
allyl-2-carbomethoxy-cyclohexanols (6d) and (7d).
Prepared from reduction of (3d); IR (film): n O–H 3457,
n CvC–H 3077, n C–H 2940 and 2863, n CvO 1727, n
C–O 1223 cm21; 1H NMR (200 MHz): 1.15–1.95 (m, 8H,
CH2 in cyclohexane ring), 2.40 (dd, 1H, J¼7.1, 14.2 Hz,
CHHCHvCH2), 2.55 (dd, 1H, J¼7.1, 14.2 Hz,
CHHCHvCH2), 3.45 (dd, 0.7H, J¼3.5, 9.8 Hz, CH(OH),
cis diastereomer), 3.69 (s, 1H, COOCH3), 3.71 (s, 2H,
COOCH3), 4.93 (dd, 0.3H, J¼3.5, 8.4 Hz, CH(OH), trans
diastereomer), 5.00–5.15 (m, 2H, CH2CHvCH2), 5.65–
5.90 (m. 1H, CH2CHvCH2) ppm; 13C NMR (50 MHz):
177.2 and 177.3 (COOCH3), 134.1 (CH2CHvCH2, trans
diastereomer), 133.3 (CH2CHvCH2, cis diastereomer),
118.4 (CH2CHvCH2, cis diastereomer), 117.7 (CH2-
CHvCH2, trans diastereomer), 74.3 (CHOH, cis dia-
stereomer), 71.5 (CHOH, trans diastereomer), 51.7
(COOCH3), 52.2 and 51.3 (C-2), 41.2 and 35.5 (CH2-
CHvCH2), 32.3 and 31.5 (CH2-6), 29.1 and 28.9 (CH2-3),
23.9 and 22.6 (CH2-5), 22.6 and 20.1 (CH2-4) ppm. Anal.
calcd for C11H18O3: C 66.64; H 9.15. Found: C 66.51; H
9.21.
4.7. Reduction of the 2-allyl-2-carboethoxy-cyclohexa-
none (3e) with zinc borohydride
To a solution of b-ketoester derivative (3e) (1 mmol) in
anhydrous THF (6 mL), cooled at 0 8C under nitrogen
atmosphere, was slowly added 2.3 mL (1.2 mmol) of
0.12 M solution42 of zinc borohydride in anhydrous THF.
The reaction mixture was stirred for 30 min at 0 8C, and
afterwards the solvent was concentrated at reduced pressure
(80 mm Hg). for 15 min at 0 8C, 5 mL of 1 M solution of
H2O2 and 7 mL of 0.2 N aq. solution of NaOH were added).
The white doughy residue was diluted with methylene
chloride (30 mL) and the resulting solution was washed with
saturated aqueous ammonium chloride solution (30 mL).
Usual workup of the organic layer afforded the diaster-
eomeric mixture of cyclohexanols (6e) and (7e) as described
in Table 3.
4.9.2. Diastereomeric mixture of trans- and cis-(6)-2-
allyl-2-carboethoxy-cyclohexanols (6e) and (7e).
Prepared from reduction of (3e); IR (film): n O–H 3490,
n CvC–H 3077, n C–H 2979, 2937 and 2863, n CvO
1
1723, n C–O 1221 and 1201 cm21; H NMR (200 MHz):
1.27 (t, 3H, J¼7.1 Hz, COOCH2CH3), 1.38–1.77 (m, 8H,
CH2 in cyclohexane ring), 2.35 (dd, 1H, J¼7.5, 14.2 Hz,
CHHCHvCH2), 2.60 (dd, 1H, J¼7.5, 14.2 Hz,
CHHCHvCH2), 3.90–4.02 (m, 1H, CH(OH)), 4.16 (q,
2H, J¼7.1 Hz, COOCH2CH3), 5.02–5.10 (m, 2H, CH2-
CHvCH2), 5.70–5.79 (m. 1H, CH2CHvCH2) ppm; 13C
NMR (50 MHz): 176.7 and 177.3 (COOCH2CH3), 134.1
(CH2CHvCH2, trans diastereomer), 133.3 (CH2CHvCH2,
cis diastereomer), 118.2 (CH2CHvCH2, cis diastereomer),
117.5 (CH2CHvCH2, trans diastereomer), 74.3 (CHOH,
cis diastereomer), 71.2 (CHOH, trans diastereomer), 60.5
(COOCH2CH3), 51.7 (C-2), 41.2 (CH2CHvCH2, cis
diastereomer), 35.7 (CH2CHvCH2, trans diastereomer),
32.2 (CH2-6, cis diastereomer), 31.5 (CH2-3, cis dia-
stereomer), 29.6 (CH2-6, trans diastereomer), 29.4 (CH2-
3, trans diastereomer), 23.9 (CH2-5, cis diastereomer),
22.5 (CH2-4, cis diastereomer), 22.3 (CH2-5, trans
4.8. Reduction of the 2-allyl-2-carboethoxy-cyclohexa-
none (3e) with lithium-tri-sec-butyl-borohydride
(L-Selectride)
To a solution of b-ketoester derivative (3e) (1 mmol) in
anhydrous THF (6 mL), cooled at 278 8C under nitrogen
atmosphere, was slowly added 1.2 mL (1.2 mmol) of 1 M
solution of lithium-tri-sec-butyl-borohydride in anhydrous
THF. The reaction mixture was stirred for 30 min at
278 8C, then for 15 min at 0 8C, and afterwards 5 mL of
1 M solution of H2O2 and 7 mL of 0.2 N aq. solution of
NaOH were added. After 15 min, the system was diluted
with ethyl ether (10 mL) and the organic layer was
separated, washed with a saturated aqueous solution of
sodium bissulfite (5 mL) and submitted to the usual workup

2754
C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
diastereomer), 21.4 (CH2-4, trans diastereomer), 14.2
(COOCH2CH3) ppm. Anal. calcd for C12H20O3: C 67.89;
H 9.50. Found: C 67.74; H 9.60.
(COOCH2CH3) ppm. Anal. calcd for C16H22O3: C 73.25; H
8.45. Found: C 73.18; H 8.37.
4.9.3. Diastereomeric mixture of trans- and cis-(6)-2-
allyl-2-carboisobutoxy-cyclohexanols (6f) and (7f).
Prepared from reduction of (3f); IR (film): n O–H 3497, n
CvC–H 3077, n C–H 2938 and 2872, n CvO 1723, n
Acknowledgements
We are grateful to CNPq (BR.), CAPES (BR.) FINEP (BR.)
and FAPERJ (BR.) for financial support and also for
fellowships (to L. H. P. T., C. M. S. M., F. R. A. Q., E. J. B.
and C. A. M. F.). We indebted to Dr. Andrew E. Greene
´
(Universite Joseph Fourier, Grenoble, France) for sugges-
tions and for a gift of tetrabutylammonium borohydride.
1
C–O 1222 and 1200 cm21; H NMR (200 MHz): 0.95 (d,
6H, J¼6.6 Hz, COOCH2CH(CH3)2), 1.25–1.96 (m, 9H,
CH2 in cyclohexane ring and COOCH2CH(CH3)2), 2.35
(dd, 1H, J¼7.1, 14.2 Hz, CHHCHvCH2), 2.60 (dd, 1H,
J¼7.1, 14.2 Hz, CHHCHvCH2), 2.90 (s, D2O exchange-
able, 1H, CH(OH)), 3.44 (dt, 0.7H, J¼9.9, 3.3 Hz, CH(OH),
cis diastereomer), 3.57 (d, 0.3H, J¼10.1 Hz, trans dia-
stereomer), 3.82–3.97 (m, 2H, COOCH2CH(CH3)), 5.05 (d,
2H, J¼12.8 Hz, CH2CHvCH2), 5.79 (qt, 1H, J¼9.5 Hz,
CH2CHvCH2) ppm; 13C NMR (50 MHz): 176.9 and 176.7
(COOCH2CH(CH3)2), 134.1 and 133.3 (CH2CHvCH2),
118.3 and 117.6 (CH2CHvCH2), 77.8 and 74.3 (CHOH),
70.9 (COOCH2CH(CH3)2), 52.1 and 51.1 (C-2), 42.0
(CH2CHvCH2), 33.0 and 32.0 (CH2-6), 29.7 (CH2-3),
27.7 (COOCH2CH(CH3)2), 24.4 and 23.0 (CH2-5), 23.0 and
22.0 (CH2-4), 19.2 and 19.1 (COOCH2CH(CH3)2) ppm.
Anal. calcd for C14H24O3: C 69.96; H 10.07. Found: C
70.09; H 10.12.
References and notes
1. Benetti, S.; Romagnoli, R.; De Risi, C.; Spalluto, G.; Zanirato,
V. Chem Rev. 1995, 95, 1065–1114.
2. (a) Ghosh, A. K.; Thompson, W. J.; Munson, P. M.; Liu, W.;
Huff, J. Bioorg. Med. Chem. Lett. 1995, 5, 83–88. (b) Cossy,
J.; Ibhi, S.; Kahn, P.; Tacchini, L. Tetrahedron Lett. 1995, 36,
7877–7880. (c) Keppens, M.; De Kimpe, M. J. Org. Chem.
1995, 60, 3916–3918.
3. Fraga, C. A. M.; Barreiro, E. J. Chirality 1996, 8, 305–310.
4. Fraga, C. A. M.; Barreiro, E. J.; da Silva, E. F.; dos Santos,
A. R.; Ramos, M. C. K. V.; de Aquino Neto, F. R. Chirality
1997, 9, 321–324.
4.9.4. Diastereomeric mixture of trans- and cis-(6)-2-
propyl-2-carboethoxy-cyclohexanols (6g) and (7g).
Prepared from reduction of (3g); IR (film): n O–H 3477,
n CvC–H 3077, n C–H 2955 and 2865, n CvO 1724, n
5. Fraga, C. A. M.; Barreiro, E. J. Synth. Commun. 1995, 25,
1133–1144.
C–O 1218 cm21 1H NMR (200 MHz): 0.86 (m, 3H,
;
6. Sant’Anna, C. M. R.; Alencastro, R. B.; Barreiro, E. J.; Fraga,
C. A. J. Mol. Struct. (Theochem) 1995, 340, 193–199.
7. Teixeira, L. H. P.; Barreiro, E. J.; Fraga, C. A. M. Synth.
Commun. 1997, 27, 3241–3257.
CH2CH2CH3), 1.23–2.10 (m, 15H, CH2 in cyclohexane
ring, CH2CH2CH3 and COOCH2CH3), 3.40 (m, 1H, CHOH,
trans diastereomer), 3.90 (m, 1H, CHOH, cis diastereomer),
4.16 (q, 2H, J¼7.0 Hz, COOCH2CH3) ppm; 13C NMR
(50 MHz): 175.8 (COOCH2CH3), 74.9 (CHOH, cis dia-
stereomer), 72.0 (CHOH, trans diastereomer), 60.6 and 60.5
(COOCH2CH3), 50.6 (C-2), 32.8 (CH2CH2CH3), 32.5
(CH2-6, cis diastereomer), 31.7 (CH2-6, trans dia-
stereomer), 29.8 (CH2-3, cis diastereomer), 29.5 (CH2-3,
trans diastereomer), 24.0 (CH2-5, cis diastereomer), 22.8
(CH2-5, trans diastereomer), 21.4 (CH2-4), 17.6 and 17.5
(CH2CH2CH3), 14.9 and 14.8 (COOCH2CH3), 14.4 (CH2-
CH2CH3) ppm. Anal. calcd for C12H12O3: C 67.26; H 10.35.
Found: C 67.33; H 10.29.
8. Teixeira, L. H. P.; de Souza, M. C. B. V.; Ramos, M. C. K. V.;
de Aquino Neto, F. R.; Barreiro, E. J.; Fraga, C. A. M. Synth.
Commun. 2002, 32, 505–526.
9. Pec¸anha, E. P.; Fraga, C. A. M.; Barreiro, E. J. Heterocycles
1998, 48, 2621–2630.
10. Barreiro, E. J.; Fraga, C. A. M.; Pec¸anha, E. P. Trends
Heterocycl. Chem. 1999, 6, 37–48.
11. Santos, M. R. L.; Barreiro, E. J.; Braz-Filho, R.; Fraga, C. A.
M. Tetrahedron 2000, 56, 5289–5295.
12. Pec¸anha, E. P.; Verli, H.; Rodrigues, C. R.; Barreiro, E. J.;
Fraga, C. A. M. Tetrahderon Lett. 2002, 43, 1607–1611.
13. Fraga, C. A. M.; Miranda, A. L. P.; Barreiro, E. J. Chem.
Pharm. Bull. 1996, 44, 2157–2161.
4.9.5. Diastereomeric mixture of trans- and cis-(6)-2-
benzyl-2-carboethoxy-cyclohexanols (6h) and (7h).
Prepared from reduction of (3h); IR (film): n O–H 3479,
n CvC–H 3063 and 3028, n C–H 2979 and 2918, n CvO
14. Pec¸anha, E. P.; Fraga, C. A. M.; de Sant’anna, C. M. R.;
Miranda, A. L. P.; Barreiro, E. J. Il Farmaco. 1998, 53,
327–336.
1
1747, n C–O 1180 cm21; H NMR (200 MHz): 1.07 (t,
´
15. Pec¸anha, E. P.; Fraga, C. A. M.; Barreiro, E. J. Quım. Nova
1997, 20, 435–437.
0.9H, J¼7.1 Hz, COOCH2CH3), 1.21 (t, 2.1H, J¼7.1 Hz,
COOCH2CH3), 1.27–2.20 (m, 5H OvCCHHCH2CH2-
CH2), 2.37–2.47 (m, 3H, OvCCHHCH2CHCH2), 2.85 (d,
0.5H, J¼14.4 Hz, PhCHH), 3.06 (s, 1.5H, PhCHH), 3.96–
4.19 (m, 2H, COOCH2CH3), 7.08–7.12 (m, 2H, metaAr-H),
7.10–7.30 (m, 5H, C6H5) ppm; 13C NMR (50 MHz): 177.3
(COOCH2CH3), 130.8 (ipsoAr), 130.1 (orthoAr), 128.1
(metaAr), 126.7 and 126.5 (paraAr), 73.7 and 71.4
(CHOH), 60.7 and 60.6 (COOCH2CH3), 53.0 (C-2), 42.4
(–CH2Ph), 32.9 and 31.8 (CH2-6), 29.8 and 28.7 (CH2-3),
24.5 and 22.9 (CH2-5), 22.4 and 21.8 (CH2-4), 14.1 and 14.0
16. Purchased from Aldrich Co, Milwaukee, USA.
17. Micovic, V. M. Organic synthesis; 1943; Collect. Vol. 2.
pp 264–265.
´ ´
18. Araujo, A. C. V.; Almeida, F. V.; Bieber, L. W. Quım. Nova
1996, 19, 79–81.
19. Barco, A.; Benetti, S.; Pollini, G. P. Synthesis 1973, 316.
20. Ramos, M. C. K. V.; Silva, E. F.; Aquino Neto, F. R.; Pec¸anha,
E. P.; Rodrigues, C. R.; Barreiro, E. J.; Fraga, C. A. M. Anal.
Chem. 2000, 72, 3056–3062.
21. Ramos, M. C. K. V.; Teixeira, L. H. P.; Aquino Neto, F. R.;

C. A. M. Fraga et al. / Tetrahedron 60 (2004) 2745–2755
2755
Barreiro, E. J.; Rodrigues, C. R.; Fraga, C. A. M.
J. Chromatogr. A 2003, 985, 321–331.
22. Frater, G. Helv. Chim. Acta 1980, 63, 1383.
31. Brown, H. C.; Krishanamurthy, S. J. Am. Chem. Soc. 1972, 94,
7159–7161.
32. Raber, D.; Guida, W. C. J. Org. Chem. 1976, 41, 690–696.
33. Akita, H.; Chen, C. Y.; Uchida, K. Tetrahedron: Asymmetry
1995, 2131–2134.
23. (a) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209–220.
(b) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 221–264.
(c) Stewart, J. J. P. J. Comput. Chem. 1991, 12, 320–341.
24. Hehre, W. J.; Shusterman, A. J.; Nelson, J. E. The molecular
modeling workbook for organic chemistry; Wavefunction:
Irvine, 1998; p 307.
34. Gung, B. W. Chem. Rev. 1999, 99, 1377–1386.
35. Cimarelli, C.; Palmieri, G. Tetrahedron: Asymmetry 2000, 11,
2555–2563.
36. Hamed, O.; El-Qisairi, A.; Henry, P. M. J. Org. Chem. 2001,
66, 180–185.
¨
25. Holtje, H.-D.; Folkers, G. Molecular modeling. Basic
principles and applications; Manhold, R., Kubinyi, H.,
Timmerman, H., Eds.; VCH: Weinheim, 1997; Vol. 5, p 194.
26. Wigfield, D. C. Tetrahedron 1979, 35, 449–462.
27. Dalla, V.; Fonken, G. J.; Noyce, D. S. Tetrahedron Lett. 1999,
40, 5193–5196.
37. Tanabe, Y. Bull. Chem. Soc. Jpn 1989, 62, 1917–1924.
38. Pei, T.; Widenhoefer, R. A. Chem. Commun. 2002, 650–651,
see supplementary data.
39. Padwa, A.; Kulkarni, Y. S.; Zhang, Z. J. Org. Chem. 1990, 55,
4144–4153.
28. Koskinen, A. M. P.; Koskinen, T. M. Tetrahedron Lett. 1993,
34, 6765–6768.
40. Ando, K.; Takemasa, Y.; Tomioka, K.; Koga, K. Tetrahedron
1993, 49, 1579–1588.
29. (a) Taniguchi, M.; Fujii, H.; Oshima, K.; Utimoto, K.
Tetrahedron 1993, 49, 11169–11182. (b) Taniguchi, M.;
Fujii, H.; Oshima, K.; Utimoto, K. Tetrahedron 1995, 51,
679–686.
41. C-Alkylated derivative was obtained in the yields described
here after its chromatographic separation from a mixture with
the corresponding O-alkylated derivative (10%), by using a
silica gel column and hexanes/AcOEt (from 99:1 to 95:5) as
eluent.
30. (a) Nakata, T.; Oishi, T. Tetrahedron Lett. 1980, 21,
1641–1644. (b) Nakata, T.; Tanaka, T.; Oishi, T. Tetrahedron
Lett. 1980, 22, 4723–4726.
´ ´
42. Crabbe, P.; Garcia, G. A.; Rıus, C. J. Chem. Soc., Perkin
Trans. 1 1973, 810–816.