α-Fluoro decalones as chiral epoxidation catalysts: Fluorine effect

Article abstract of DOI:10.1016/j.tetasy.2004.10.033

Three rigid monofluorinated trans-decalones 4a, 5e, and 6e (90% ee) have been synthesized from commercially available (-)-(R)-methyl naphthalenone (90% ee). Their structures have been fully characterized (NMR, X-ray): ketones 4a and 5e are Me,F-disubstituted α to the carbonyl with the fluorine axial and equatorial, respectively, while ketone 6e is F-monosubstituted α to the carbonyl with the fluorine equatorial. The use of these ketones as chiral catalysts for the epoxidation of trans-olefins (such as stilbene, β-methylstyrene and p-methoxy cinnamate) through the formation of dioxiranes shows (i) that dioxiranes with an equatorial fluorine α to the dioxirane ring are less reactive and provide lower ee's than dioxiranes with an axial fluorine and having the same chirality and (ii) that an axial methyl α to the dioxirane ring is significantly less efficient than a fluorine. The results corroborate Armstrong and Houk's theoretical model and our first hypothesis to rationalize the inverted enantioselectivities observed using α-fluorinated cyclohexanones having the same chirality, i.e.: rapid ring inversion (Curtin-Hammett principle) allows the dioxirane conformation to have the fluorine axial (even if less populated than the other) to contribute significantly to the epoxidation reaction.

Full text of DOI:10.1016/j.tetasy.2004.10.033

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 3891–3898
a-Fluoro decalones as chiral epoxidation catalysts: fluorine effect
a,
Arlette Solladie-Cavallo, Loıc Jierry, Arlette Klein,
¨
a
a
*
´
Michel Schmitta and Richard Welterb
a
´ ´ ´ ´
Laboratoire de Stereochimie Organometallique associe au CNRS, ECPM/Universite L.Pasteur, 25 rue Becquerel,
67087 Strasbourg, France
´
b
´
Laboratoire DECMET, UMR 7513, Universite Louis Pasteur, 67000 Strasbourg, France
Received 22 September 2004; accepted 27 October 2004
Available online 24 November 2004
Abstract—Three rigid monofluorinated trans-decalones 4a, 5e, and 6e (90% ee) have been synthesized from commercially available
(ꢀ)-(R)-methyl naphthalenone (90% ee). Their structures have been fully characterized (NMR, X-ray): ketones 4a and 5e are Me,F-
disubstituted a to the carbonyl with the fluorine axial and equatorial, respectively, while ketone 6e is F-monosubstituted a to the
carbonyl with the fluorine equatorial. The use of these ketones as chiral catalysts for the epoxidation of trans-olefins (such as stil-
bene, b-methylstyrene and p-methoxy cinnamate) through the formation of dioxiranes shows (i) that dioxiranes with an equatorial
fluorine a to the dioxirane ring are less reactive and provide lower eeꢀs than dioxiranes with an axial fluorine and having the same
chirality and (ii) that an axial methyl a to the dioxirane ring is significantly less efficient than a fluorine. The results corroborate
Armstrong and Houkꢀs theoretical model and our first hypothesis to rationalize the inverted enantioselectivities observed using
a-fluorinated cyclohexanones having the same chirality, i.e.: rapid ring inversion (Curtin–Hammett principle) allows the dioxirane
conformation to have the fluorine axial (even if less populated than the other) to contribute significantly to the epoxidation reaction.
ꢀ 2004 Published by Elsevier Ltd.
1. Introduction
chiral dioxiranes generated in situ from tri- and
tetra-substituted a-fluoro cyclohexanones it was found
that
Since the first use by Curci et al. in 19891 of trifluoro-
methyl acetone as a dioxirane precursor for the epoxida-
tion of olefins, the observed activating effect of fluorine
substitution has been extended to asymmetric epoxida-
tion while various chiral fluorinated ketones have been
designed.2–9
• (2S,5R)-ketones, called F-axial, were more efficient,5,6
providing epoxides with higher yields and higher ees
than (2R,5R)-ketones, called F-equatorial, (e.g.,
ketone 1a vs ketone 1e). It is noteworthy that ketones
are named F-axial and/or F-equatorial for conven-
ience and according to the NMR data (in benzene-
d6), which fit with F-axial or F-equatorial.
During work on the epoxidation of trans-methyl
p-methoxy cinnamate10 and other trans-olefins using
Me
O
F
F
Me
Me
Me
Me
F
F
F
O
O
O
O
2
5
2
5
Ph
Me
F
F
1e
F-equatorial
3e
2a
F-axial
3a
F-axial
1a
F-axial
F-equatorial
stilbene
stilbene
stilbene
stilbene
stilbene
95% conv.
78% conv.
90%conv.
69% conv.
70% conv.
90% ee (-)-(S,S) 60% ee (-)-(S,S)
90% ee(-)-(S,S)
18% ee (+)-(R,R)
20% ee (+)-(R,R)
*
Corresponding author. Fax: +33 3 90 24 27 06; e-mail: cavallo@chimie.u-strasbg.fr
0957-4166/$ - see front matter ꢀ 2004 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2004.10.033

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3892
• Di-substitution at C5 increased the enantioselectivity
by 25–30%,9 suggesting that axial approaches of the
olefin towards the corresponding dioxirane could
contribute significantly (e.g., compare ketones 2a
and 3a).
stilbene, trans-b-methylstyrene and trans-methyl p-
methoxycinnamate. The trans-decalones 5e and 6e have
been envisaged because we did not succeed in the syn-
thesis of isomer 4e.
• Inversions of enantioselectivity were observed in the
case of dioxiranes derived from (2R,5R)-ketones (F-
equatorial). This behavior was ascribed to the contri-
bution of both conformers C1 (providing enantiomer
1) and C2 (providing enantiomer 2) to the reaction
due to rapid ring inversion in the dioxiranes (Scheme
1), and to better reactivity of C2.6
2. Results and discussion
2.1. Synthesis
As shown in Scheme 2, the desired ketone 4a was
obtained in seven steps as a single diastereomer from
commercially available (ꢀ)-(R)-4,4a,5,6,7,8-hexahydro-
4a-methyl-2(3H)-naphthalenone 7. The starting com-
pound 7 was assigned an enantiomeric excess of
Rigid trans-decalones 4a, 5e, and 6e, whose dioxiranes
do not undergo chair–chair ring inversion, have thus
been envisaged to check the role of ring inversion and
of fluorine orientation on the efficiency and enantio-
selectivity of epoxidation reactions.
90% based on the measured specific rotation {[a]D
=
ꢀ198 (c 1.1, MeOH)} compared with the maximum
specific rotation {[a]D Max = ꢀ219 (c 1.1, MeOH)}
obtained by Toda and Tanaka11 after resolution with
TADDOL.
Li/NH3 reduction12 of enone 7 provided 95% of a mix-
ture of 8 (90%) and of the corresponding decalol
(10%), which are separated by chromatography. 1H
and 13C NMR of 8 were consistent with Vecchi et al.ꢀs
results13 and with a trans relationshipbetween the rings.
Me
Me
O
O
F
Me
Me
F
4a
4e
not obtained
Upon addition of MeLi,14 the two expected diastereo-
mers 9I + 9II were obtained as a 61/39 mixture, in
95% yield. The mixture was used as such for the next
dehydration step14 leading, in quantitative yield, to a
mixture of 10 (90%) and 11 (10%). Hydroboration of
the 10 and 11 mixture provided a 95/5 mixture of
trans-12 (95%) and trans-13 (5%) in 76% yield.
Me
Me
Me
F
F
O
O
5e
6e
We herein report the synthesis of ketones 4a, 5e, and 6e
from commercially available (and 90% enantioenriched)
(ꢀ)-(R)-4,4a,5,6,7,8-hexahydro-4a-methyl-2(3H)-naphth-
alenone 7 (Scheme 2), their structures (X-ray, NMR)
and their use as catalysts for the epoxidation of trans-
The desired methyl decalone 14 was then obtained
through PCC oxidation in 62% isolated yield as a single
diastereomer and without traces of the ketone corre-
sponding to 13. The thermodynamic silylenolate 15
was obtained in quantitative yield after modification of
O
O
F
Me
F
Me
Me
F
O
O
O
O
R
R
R
Me
C1
0.54 kcal/mol
O
F
R
O
C1*
C2*
0.0 kcal/mol
C2
0.0 kcal/mol
7.04 kcal/mol
oxone
oxone
Me
F
Me
F
O
O
Me
O
F
R
R
O
Me
F
R
R
K1**
K1**
K2
K2
1
* most populated according to modelization, cf. ref 6. ** most populated according to H
NMR in C D .
6
6
Scheme 1.

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3893
Me
Me
Me
Me
Me
95%
b
88%
quant.
c
+
isol.
a
OH
O
O
Me
Me
Me
H
H
H
H
(-)-(R)-7
10 (90%)
11 (10%)
8
9 (I/II=61/39)
d
76%
Me
Me
Me
Me
Me
OH
O
OSiMe3
O
e
81%
g
4e
0%
+
+
quant.
f
62%
Me
Me
Me
Me
Me
H
H
15
H
14
H
H
F
OH
4a, ~100%
trans-12 (95)
13 (5)
Scheme 2. Reagents and conditions: (a) Li/NH3, Et2O, ꢀ78°C, EtOH, from ꢀ78° C to rt, NH4Cl (cf. Ref. 13); (b) MeLi/THF, reflux; (c) pTsOH
(0.02equiv)/benzene, reflux (cf. Ref. 14); (d) BH3: THF, from 0°C to rt, NaOH, H2O2; (e) PCC (2equiv), CH2Cl2, from 0°C to rt; (f) TMSCl/NaI
(2equiv), Et3N (2equiv), MeCN/pentane, rt (cf. Ref. 15); (g) Selectfluor (1.4equiv)/DMF, 0°C.
Me
Me
F
+
20e (5%)
6e (95%)
O
O
F
c1
Me
80%
Me
Me
Me
Me
Me
Me
b
a
Me
Me
O
quant.
78%
F
O
O
OSiR3
OSiR3
O
(-)-(R)-7
18
+
c2
85%
19
+
8
d
16 (87%)
5e (92%)
quant.
+
80%
e
+
Me
20e (8%)
+ 6e (0%)
16+17
8
OSiR3
17 (13%)
Scheme 3. Reagents and conditions: (a) Li/NH3, Et2O, ꢀ78°C, EtOH, from ꢀ78°C to rt, NH4Cl (cf. Ref. 13); (b) TMSCl/NaI (2equiv), NEt3
(2equiv), MeCN/pentane, rt (cf. Ref. 15); (c1) Selectfluor (1.4equiv)/DMF, 0°C; (c2) MeI/TBAF/THF; (d) TMSCl/NaI (2equiv), NEt3 (2equiv),
MeCN/pentane, rt (cf. Ref. 15); (e) Selectfluor (1.4equiv)/DMF, 0°C.
the literature work-up15 while the fluorination was per-
formed with Selectfluor in 81% yield providing a single
diastereomer identified (cf. below) as 4a (no traces of
4e were detected by NMR). The high diastereoselectivity
of this stepcould be due to the better availability of the
face of the double bond trans to the methyl groupin the
twist conformation of the ring. Diastereomer 4a was
thus obtained in 32% overall yield.
2.2. Structural determination of 4a, 5e, and 6e
The trans-structure, chair conformation and axial orien-
tation of the fluorine in ketone 4a were determined by
X-ray analysis of a single crystal (Fig. 1).
Ketone 5e was synthesized in five steps from (ꢀ)-(R)-7
(Scheme 3) and obtained as a single diastereomer (iso-
mer 5a has not been detected) although traces (8%) of
ketone 20e was obtained simultaneously because of the
formation of silylenolate 17, which seemingly did not
undergo methylation during step3.
Ketone 6e was synthesized in three steps from (ꢀ)-(R)-7
(Scheme 3) and obtained as a 95/5 mixture of 6e (major,
95%) and ketone 20e (5%), due again to the presence of
silylenolate 17.
After chromatographic purification, ketone 5e was ob-
tained pure in 48% overall yield, while 6e was obtained
as a 95/5 mixture of 6e and 20e (both having the fluorine
equatorial) in 56% overall yield.
Figure 1. ORTEP plot of one molecule of 4a: F–C2–C1–O = ꢀ117.5°;
F–C2–C1–C12 = 64.2°. Configuration (2R,5R,10R) from the known
(R)-configuration at ClO.

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3894
Therefore, a-fluoro decalone 4a was assigned the
(2R,5R,10R)-configuration (according to the X-ray
numbering of atoms shown in Fig. 1) on the basis of
the known (R)-configuration at C10 of the starting
decalenone 7 (Fig. 1). NOESY (Fig. 2) and the 6.5Hz
value for JHF corroborated the trans-(R,R)-structure
of the decalone 4a and the axial orientation of the fluo-
rine atom.
2.3. Epoxidation of trans-olefins
The results of the asymmetric epoxidation of stilbene 21,
b-methylstyrene 22 and methyl p-methoxy cinnamate 23
(Scheme 4) with these ketones are shown in Table 1. The
conversions have been determined by combining weights
and 1H NMR (300 and/or 400MHz) of solvent-free
crude products. Epoxides were then isolated by flash
chromatography on silica gel. The ees of the epoxides
4
16
1
were determined by H NMR using Eu(hfc)3 in CDCl3
and absolute configurations determined by comparing
the measured specific rotation with the literature val-
ues.17 The ees (Table 1, columns 7 and 9) have been cor-
rected considering that all ketones used were 90% ee.
(R)
NOE
He
O
Me
4a
4
Me
2
Ha
(R)
F
R1
H
H
R1
H
O
H4e = doublet, 2J(H4e-H4a)= 12.5 Hz and NOE with Me
H
H4a = double doublet, 2J(H4a-H4e) = 12.5 Hz and 4JHF = 6.5 Hz
2
3
Figure 2.
ketone (0.3 equiv.)
+ oxone
R
R
21: R=H; R1=Ph
22: R=H; R1=Me
24 (+)-(R,R)
25: (+)-(R,R)
26: (+)-(2S,3R)
a-Fluoro decalones 5e and 6e were assigned the
(3S,R,R)-configuration (according to Fig. 3 numbering)
with an equatorial fluorine (Fig. 3) using NOESY and
1H NMR patterns.
23: R=MeO; R1=CO2Me
Scheme 4.
5e
6e
NOE
NOE
(R)
(R)
- Me/Me NOE
Ha
- Me/H3a NOE
Me
He
- H1a = d.d : 2J(H1e-H1a)= 15 Hz ;
- H3a = d.d.d : 2J(H3a-F) = 48 Hz ;
3J(H3a-H4a) (180˚) = 12.5 Hz;
3J(H3a-H4e) (60˚) = 6.5 Hz;
Me
Ha
Me
3
3J(H1a-H9a) (180˚) = 13 Hz; no 4JH1aF
4
F
3
2
F
- H1e = d.d.d : J(H1e-H1a)= 15 Hz ;
3J(H1e-H9a) (60˚) = 3 Hz; JH1eF=5Hz
4
O
1
Ha
O
(R)
He
Ha
(R)
Figure 3.
Table 1. Epoxidations of stilbene 21, b-methylstyrene 22 and methyl p-methoxy cinnamate 23
Me
O
O
O
O
Me
Me
H
H
O
O
Me
F
F
Me
F
H
Dioxiranes from :
4a
5e
6e
Olefin
Ketone
Solventa
Temp. (°C)
React. time (h)
Conv. (%)b
Er (%)c
Absol. Conf.d
Ee (%)
PhCH = CHMe
4a
5e
6e
Diox/H2O
Diox/H2O
Diox/H2O
25
25
25
6
6
6
100
88
85/15
61/39
50/50
(+)-(R,R)
(+)-(R,R)
—
70
22
0
100
PhCH = CHPh
Cinnamate
4a
5e
6e
Diox/H2O
Diox/H2O
Diox/H2O
25
25
25
6
6
6
100
0
93/7
—
(+)-(R,R)
86
—
—
—
—
0
—
4a
5e
6e
Diox/H2O
Diox/H2O
Diox/H2O
25
25
25
6
6
6
75
0
73/27
—
—
(+)-(2S,R)
46
—
—
—
—
0
a Diox/H2O: 2/1 (3equiv of oxone), rt.
b Determined by 1H NMR (300 and/or 400MHz) on the crude products of the reactions.
c Enantiomeric ratios have been determined by 1H NMR (400MHz) using Eu(hfc)3 in CDCl3: the precision is 2%.
d Absolute configurations have been determined from the sign of the specific rotation as compared with literature results (cf. Ref. 17).

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3895
From previous results5 and theoretical calculation by
Armstrong et al.,18 decalones 5e and 6e with the fluorine
equatorial were expected to be weaker catalysts, than
ketone 4a with the fluorine axial and, as expected, did
not catalyze the epoxidation of stilbene 21 and cinna-
mate 23 but did catalyze the epoxidation of b-methyl
styrene 22.3,5,8
one could expect the enantiomers (+)-(R,R)-24, (+)-
(R,R)-25 and (+)-(2S,3R)-26 to be obtained, which is in-
deed observed (Table 1, lines 1, 2, 4, and 8). Equatorial
E-I approach with no nÆp repulsion (between F and the
phenyl) involved was expected to be favored over the E-
II approach, which involves such a nÆp repulsion. Simi-
larly, in the case of axial approaches A-II, which in-
volves no steric MeÆPh repulsive interaction, is favored
over A-I. Both E-I and A-II provide the same enantio-
mer of the epoxide.
Considering the configurations of 4a and 5e, using the
well-accepted spiro-model19 and both equatorial20–22
and axial9 approaches (Fig. 4 and Fig. 5 with R = Me)
Furthermore ketone 6e (95%) as well as ketone 20e (5%)
provide almost racemic epoxides (Fig. 5 with R = H)
again as observed (Table 1, line 3). In this case all
approaches (E-I, E-II, A-I, and A-II) involve HÆPh
interactions.
Ph
Ph
Ph
(+)-(R,R)
Ph
(-)-(S,S)
O
O
Ph
Axial approaches
Ph
Ph
Me
Ph
A-II
Me
A-I
3. Conclusion
O
O
O
Although direct epoxidation of b-methyl styrene by
oxone is a competitive reaction responsible for the
lower ee observed with this olefin,23 the observation
of eeꢀs by using 4a (70%: Table 1, line 1) and 5e (22%:
Table 1, line 2) indicates that these ketones play
the expected role of a catalyst (the corresponding
dioxiranes ?tul?> providing at least part of the epoxida-
tion reaction). It can be suggested that this is also true
for ketone 6e, which provides part of the epoxidation
but toward a racemic mixture as expected from the
model.
O
Me
F
Me
F
E-II
E-I
Ph
Ph
Equatorial approaches
Ph
Ph
O
O
(-)-(S,S)
(+)-(R,R)
Ph
Ph
Ph
Ph
E-I >> E-II
A-II > A-I
Me/Ph repulsion in A-I
F/Ph n.π repulsion in E-II
In summary, as predicted from the model and consist-
ent with Armstrong and Houkꢀs theoretical calcu-
lations:18
Figure 4. Model epoxidation of trans-stilbene by the dioxirane derived
from 4a.
Ph
Ph
Ph
(+)-(R,R)
Ph
(-)-(S,S)
O
O
Ph
Axial- approaches
Ph
Ph
Ph
A-II
A-I
O
O
Me
Me
H
R
H
R
O
O
F
F
E-II
E-I
Ph
Ph
Equatorial- approaches
Ph
Ph
O
O
(-)-(S,S)
(+)-(R,R)
Ph
Ph
Ph
Ph
E-I~E-II when R=H
H/Ph interaction in both cases
[(-)-epoxide] ~[(+)-epoxide] expected
A-I ~ A-II when R=H or Me
H/Ph interaction in both cases
[(-)-epoxide] ~[(+)-epoxide] expected
E-I >E-II when R=Me
Me/Ph repulsion in E-II
[(+)-epoxide] > [(-)-epoxide] expected
Figure 5. Model epoxidation of trans-stilbene by the dioxirane derived from 5e (R = Me) and 6e (R = H).

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3896
• Dioxiranes with an axial a-fluorine (and an equato-
rial methyl) are more reactive and provide higher
ees than dioxiranes with an axial methyl (and an
equatorial fluorine): Table 1, compare line 1 with line
2 and lines 4 and 7 with lines 5 and 8.
orga. Reactions were monitored by TLC using Merckꢀs
glass plates with silica gel 60 F254. Silica gel Si 60 (40–
60lm) from Merck was used for the chromatographic
purifications.
• An axial methyl group(ketone 5e) is less efficient than
an axial fluorine (ketone 4a) providing lower yields
and lower ees (considering, of course, that the extent
of direct epoxidation is identical/similar in both
cases): Table 1, compare line 1 with line 2.
• An axial proton even with an a-fluorine equatorial
has no effect on the enantioselectivity: Table 1, com-
pare line 1 with line 3.
4.1. Silylenol ether: general procedure
To a solution of the desired ketone (1equiv) in dry pen-
tane (7mL for 9mmol) were added, under argon, Et3N
(2equiv) and TMSCl (2equiv) followed by a solution
of anhydrous NaI (2equiv) in dry CH3CN (20mL).12
After 3 or 4h at room temperature, stirring was stopped
and the upper organic layer (pentane) transferred into a
dry flask. The remaining mixture was extracted with dry
pentane until no trace of silylenol ether was detected on
TLC in the acetonitrile layer.15 The pentane phases were
gathered and evaporated to give a colorless oil, pure by
NMR analysis.
It thus appears that our very first hypothesis6 to ration-
alize the inverted enantioselectivities and high yields ob-
tained during epoxidation of stilbene and/or cinnamate
using (2R,5R)-cyclohexanones
satisfactory.
(F-equatorial)
is
4.2. Fluoration: general procedure
1. The Curtin–Hammett principle holds true because of
rapid exchange between the dioxirane ring conforma-
tions C1 (with an equatorial fluorine) and C2 (with an
axial fluorine).
2. Inversion of the ring is equivalent to an inversion of
configuration. Therefore, if C1 provides the (ꢀ)-
enantiomer, C2 will provide (+)-enantiomer.
3. C2, in which the fluorine is axial, has a higher
reactivity18 than C1 and could/will contribute to the
reaction even if less populated than C1, which will
increase the apparent reactivity and decrease the
enantioselectivity.
To a solution of the desired silylenol ether (1equiv) in
dry DMF (25mL for 7mmol) was added dropwise,
under argon, a solution of Selecfluor (1.4equiv) in
DMF (40mL). The mixture was stirred at 0°C for 2h.
Then water (50mL) was poured into the reaction mix-
ture, the different phases were separated and the aque-
ous phase was extracted with ether (3 · 100mL). The
organic phases dried over MgSO4. After filtration and
evaporation, the residue (containing DMF) was directly
purified by column chromatography with an appropri-
ate pentane/ether gradient.
4. Calculation and modeling show that conformation
C2 with an F-axial is significantly populated, which
makes its contribution larger.
4.3. Epoxidation: general procedure
5. The enantioselectivity observed is the result of three
rate constants [rate of exchange, reactivity of C1 pro-
viding the (ꢀ)-enantiomer and of C2 providing the
(+)-enantiomer].
6. If the contribution of C2 to the reaction is large
enough, the (+)-enantiomer will dominate and an
inversion of enantioselectivity will be observed.
Distilled water (6mL) and a solution (4mL) of acetic
acid (0.5mL) and 0.1M aqueous K2CO3 (100mL) were
added under stirring to a solution of 1mmol of the
desired olefin and 0.1 or 0.3mmol of ketone (0.1 or
0.3equiv) in DME (16mL). The mixture was maintained
at the desired temperature and a solution of oxone
(1.850g, 3mmol, 6equiv of oxidant) in distilled water
(7mL) added dropwise over 6h. During the addition
of oxone, the pH was controlled and regulated (ꢁ8.5–
9) by the addition of a solution of K2CO3 1M. The reac-
tion was immediately quenched by the addition of
CH2Cl2 (30mL) and water (10mL). The mixture was ex-
tracted with CH2Cl2 (3 · 20mL), dried over Na2SO4 and
analyzed by NMR. After chromatographic purification,
the ketone was recovered and the isolated epoxide ana-
lyzed (NMR, optical rotation and/or chiral HPLC).
4. Experimental
1H and 13C NMR spectra were recorded on a Bruker
AC 300 (300MHz) or a Bruker Avance (400MHz) spec-
trometer with CDCl3 or C6D6 as solvent. Chemical
shifts (d) are given in ppm downfield from TMS. Optical
rotations were determined with a Perkin–Elmer 341
polarimeter. Rotations were determined in EtOH
20
D
(c = 1) for stilbene oxide {lit.17; (S,S): ½aŠ ¼ ꢀ299,
4.4. trans-10-Methyl decal-2-one, 8
ꢁ100% ee}, in MeOH (c = 0.5) for the methyl p-
20
D
1
methoxy cinnamate epoxide {lit.17; (2R,3S): ½aŠ
¼
For H and 13C NMR of 8 see Ref. 13.
ꢀ212, ꢁ100% ee} and neat for phenylpropileneoxide
20
D
{lit.17; (R,R): ½aŠ ¼ þ110}. (ꢀ)-(R)-4,4a,5,6,7,8-
4.5. trans-1,10-Dimethyl decal-1-ol, 9
Hexahydro-4a-methyl-2(3H)-naphthalenone (90% ee)
was purchased from Aldrich. Stilbene oxide and styrene
oxide exhibit identical NMR spectra with the racemic
compounds purchased from Aldrich. Methyl p-methoxy
cinnamate epoxide exhibits identical NMR spectra with
the racemic compound provided to us by Sylachim–Fin-
2 Diastereomers I/II = 61/39. 1H NMR (CDCl3) 0.79 (s;
3H, Me, I); 0.84 (s, 3H, Me, II); 1.21 (s, 3H, Me, I); 1.26
(s, 3H, Me, II) and all ring protons between 1.1 and
1.5ppm. 13C NMR (CDCl3) I: 14.7; 22.0; 26.9; 28.6;
31.8; 33.2; 34.8; 37.3; 39.8; 41.3; 42.0; 70.4. II: 15.3;

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3897
21.9; 26.8; 26.9; 28.8; 33.7; 36.4; 39.6; 41.2; 42.7; 43.5;
71.9.
(br dt, 2H); 4.70 (m, 1H). 1H NMR (CDCl3), 17: all sig-
nals overlapped with those of 16 but 4.50 (br, 1H).
4.6. Decalene 10 (90%)
13C NMR (CDCl3), 16: 0.7; 16.5; 22.6, 26.9; 29.2; 32.4;
35.1; 40.9; 41.1; 41.2; 103.4; 149.0.
1H NMR (CDCl3) 0.77 (s, 3H, Me); 1.62 (s, 3H, Me);
ring signals 1.1–1.62; 5.27 (bm, 1H, olefinic). 13C
NMR (CDCl3) 16.8; 22.4; 23.5; 27.1; 29.0; 32.1; 35.3;
40.5; 41.3; 119.9; 133.1.
13C NMR (CDCl3), 17: 0.7; 15.3; 22.2; 27.1; 28.2; 28.4;
33.0; 38.7; 39.8; 43.7, 109.3, 149.0.
4.14. Decalone 18
4.7. Decalene 11 (10%)
1H NMR (CDCl3) 0.95 (d, 3H, 3JHH = Hz, Me); 1.0 (s,
3H, Me); 1.1–1.2 (m, 3H); 1.45 (m, 5H); 1.65 (m, 2H);
2.16 (AB part of an ABX, 2H, JHH = 14Hz,
1H NMR (CDCl3): all signals overlapped with those of
10 but 0.76 (s, 3H, Me) and 5.02 (br, 1H).
2
3JAX ꢁ 4Hz, JBX ꢁ 14Hz); 2.50 (d.pint = 7 lines,
3
3
4.8. Dimethyl decalol 12 (95%)
3JMeH = 3Jgauche = 6.5Hz, Jtrans = 13Hz, 1H). 13C
NMR (CDCl3) 14.5; 15.8; 21.4; 26.0; 28.8; 34.0; 40.2;
41.3; 45.1; 45.9; 51.5; 212.8.
3
1H NMR (CDCl3) 0.83 (s, 3H, Me); 0.93 (d, J = 6Hz,
3
3H, Me); 0.9–1.7 (m, 15H); 3.39 (ddd br, Jgauche
=
4.5Hz, Jtrans = 9.5 and 11Hz, 1H, 95%). 13C NMR
(CDCl3) 16.8; 18.6; 21.4; 26.9; 28.0; 35.3; 36.9; 41.3;
41.4; 45.0; 50.9; 73.3.
4.15. Decalone 5e
3
25
½aŠ ¼ þ84 (c 1.03, CHCl3) (90% ee). IR (CHCl3)
D
1730cmꢀ1. Anal. Calcd for C12H19FO: C, 72.68; H,
9.65. Found: C, 72.24; H, 9.11. H NMR (C6D6) 0.50
(s, 3H, Me); 0.65–1.25 (m, 8H); 1.30 (d, 3H,
1
4.9. Dimethyl decalol 13(5%)
All signals overlapped with those of 12 but 3.41 (t,
3Jtrans = 11Hz, 1H).
3JHF = 22Hz, Me); 1.40 (m, 1H); 1.52 (t, 1H,
2
2JHH = 3JHF = 14Hz); 1.71 (dd, 1H, JHH = 14Hz,
2
3JHF = 26Hz); 1.83 (dd, 1Haxial, JHH = 14.5Hz,
4.10. Dimethyl decalone 14
3JHHtrans = 13Hz);
2.25
(ddd,
1Hequatorial,
2JHH = 14.5Hz, 3JHHcis = 3Hz, 4JHF = 5Hz). 13C
NMR (CDCl3) 17.1; 20.8; 24.4 (d, JCF = 27Hz);
25
½aŠ ¼ ꢀ38 (c 1.17, CHCl3) (90% ee). 1H NMR (CDCl3)
2
D
2
225.7; 28.2, 40.8, 42.3; 43.7; 54.4 (d, JCF = 21Hz).
0.74 (s, 3H; Me); 0.99 (d, 3J = 6Hz, 3H, Me); 1.1–1.9 (m,
11H); 2.08 (d, J = 12.5Hz, 1H); 2.18 (d, J = 12.5Hz,
2
2
3
3
1H); 2.38 (qdd, J = 6.5Hz three times, Jtrans = 9Hz,
3Jgauche = 2Hz, 1H). 13C NMR (CDCl3) 14.4; 16.5;
21.2; 26.6; 27.9; 38.6; 38.8; 41.2; 44.3; 45.4; 56.9; 212.9.
4.16. Decalone 6e
25
½aŠ ¼ þ66 (c 1.10, CHCl3) (containing 5% of 20e, both
D
being 90% ee). IR (CHCl3) 1730cmꢀ1. Anal. Calcd for
C11H17FO: C, 71.70; H, 9.29. Found: C, 71.22; H,
9.01. 1H NMR (CDCl3) 1.10 (s, 3H, Me); 1.15–1.35
(m, 3H); 1.45–1.55 (m, 6H); 1.75 (m, 1H); 2.22 (m,
4.11. Trimethylsilyl derivative 15
1H NMR (CDCl3) 0.18 (s, 9H); 0.82 (s, 3H, Me); 1.55 (s,
3H, Me); 1.15–1.6 (m, 12H); 1.93 (d, J = 20Hz, 1H).
2
2
3
3H); 5.05 (ddd, 1Haxial, JHF = 48Hz, JHHtrans =
13C NMR (CDCl3) 1.15, 16.3; 16.8; 22.3; 27.1; 28.7;
34.2; 35.6; 40.9; 41.5; 47.4; 111.0; 142.0.
13Hz, JHHgauche = 7Hz). 13C NMR (C6D6) 15.7; 20.5;
25.5; 27.9; 34.1; 39.6; 43.5; 47.8 (d, JCF = 17Hz); 90.4
3
2
1
(d, JCF = 188Hz); 202.6 (d, JCF = 12Hz).
2
4.12. Decalone 4a
4.17. Decalone 20e
25
½aŠ ¼ ꢀ101 (c 1.0, CHCl3) (90% ee). IR (CHCl3)
D
1730cmꢀ1. Anal. Calcd for C12H19FO: C, 72.68; H,
9.65. Found: C, 72.32; H, 9.14. H NMR (C6D6) 0.49
1H NMR (CDCl3) signals overlapped with those of 6e
but 4.62 (dd, 1Haxial, JHF = 49Hz, JHHtrans = 12Hz).
1
2
3
(s, 3H, Me); 0.68 (q br, 1H); 0.90–1.20 (m, 6H);
1.23 (m, 1H); 1.32 (d, 3H, JHF = 22Hz, Me); 1.52
13C NMR (C6D6) 15.7; 21.0; 23.7; 25.0; 36.7; 40.2;
40.3; 50.0; 93.3 (d, JCF = 189Hz).
3
1
2
(m, 1H); 1.62 (ddd, 1H, JHH = 12.5Hz, JHH trans
3
=
13Hz, JHF = 4Hz); 1.79 (tt, 1H, Jtrans = 3Jtrans = 13
4.18. X-ray structure analysis of 4a
3
3
3
Hz, Jcis = 3Jcis = 3.5Hz); 1.95 (d, 1H, 2J = 12.5
4
Hz, JHF = 6.5Hz). 13C NMR (C6D6) 15.7; 20.5 (d,
The selected crystal was mounted on a Nonius Kappa-
CCD area detector diffractometer (Mo Ka, k =
2JCF = 24Hz); 21.2; 26.6; 27.4; 38.5; 39.2; 40.9;
42.6 (d, JCF = 22Hz); 53.2 (d, JCF = 2Hz); 96.5 (d,
2
3
˚
0.71073A). The complete conditions of data collection
2
1JCF = 169Hz); 204.7 (d, JCF = 25Hz).
(Denzo software) and structure refinements are given
below. The cell parameters were determined from
reflections taken from one set of ten frames (1.0° steps
in phi angle), each at 20s exposure. The structures were
solved using direct methods (SIR97) and refined against
F2 using the SHELXL97 software. All non-hydrogen
4.13. Trimethylsilyl derivatives 16 and 17 = 87/13
1H NMR (CDCl3), 16: 0.20 (s, 9H, Me); 0.85 (s, 3H,
Me); 1.15 (m, 4H); 1.50 (m, 5H); 1.70 (br d, 2H); 1.85

´
A.Solladie -Cavallo et al./ Tetrahedron: Asymmetry 15 (2004) 3891–3898
3898
8610; (b) Denmark, S. E.; Matsuhashi, H. J.Org.Chem.
2002, 67, 3479; (c) Stearman, C. J.; Behar, V. Tetrahedron
Lett. 2002, 43, 1943; (d) Klein, S.; Roberts, S. M. J.Chem.
Soc.Perkin Trans.1 2002, 19, 2686.
atoms were refined anisotropically. Hydrogen atoms
were generated according to stereochemistry and refined
using a riding model in SHELXL97.
´
´
9. Solladie-Cavallo, A.; Bouerat, L.; Jierry, L. Eur.J.Org.
Crystallographic data (excluding structure factors) have
been deposited in the Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC 249289.
Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax: (+44)1223-336-033; e-mail:
deposit@ccdc.cam.ac.uk).
Chem. 2001, 4557–4560.
10. The corresponding epoxide is a key intermediate in the
synthesis of hypotensive drugs, such as diltiazem, clen-
tiaezm and naltiazem.
11. Toda, F.; Tanaka, K. Tetrahedron Lett. 1988, 29, 551–554.
12. Marshall, J. A.; Cohen, N.; Arenson, K. R. J.Org.Chem.
1965, 30, 762. These authors obtained the decalol in larger
amounts and oxidized the crude product using CrO3 to
form the desired decalone. In our case the decalol was
minor and a separation was prefered.
13. Di Maio, G.; Migneco, L. M.; Vecchi, E.; Lavarone, C.
Magn.Resn.Chem. 2000, 38, 108.
14. Piers, E.; Britton, R. W.; De Waal, W. Can.J.Chem.
1969, 47, 831. For the dehydration stepthese authors
obtained only isomer 10, while we got a 9/1 mixture of 10
and 11.
Crystal data and structure refinement details: Colorless
crystal; crystal dimension: 0.22 · 0.17 · 0.10mm3;
C12H19FO, M = 198.27gmolꢀ1; orthorhombic; space
˚
˚
groupP2 12121; a = 7.479(5)A; b = 7.578(5)A; c =
19.197(5)A; Z = 4; Dc = 1.21gcmꢀ3
; l(Mo Ka) =
˚
0.086mmꢀ1; a total of 1310 reflections; 2.89° < h <
26.967°, 1310 independent reflections with 1279 having
I > 2r(I); 127 parameters; Final results: R1 = 0.0256;
15. Quantitative yield of silylenolate was obtained after
modification of the literature work-up: no addition of
wR2 = 0.0614, Goof = 1.ꢀ0340, maximum residual elec-
˚
tronic density = 0.097eA
´
´
water at all, cf.: Solladie-Cavallo, A.; Jierry, L.; Bouerat,
L.; Taillasson, P. Tetrahedron: Asymmetry 2001, 12, 883–
891. Milan Balaz, PhD dissertation, Strasbourg, June
2003.
.
Acknowledgements
´
16. (a) Solladie-Cavallo, A.; Jierry, L.; Bouerat, L.; Taillas-
son, P. Tetrahedron: Asymmetry 2001, 12, 883; (b)
´
This work was supported by the MENRT-France with a
grant to Jierry.
´
Solladie-Cavallo, A.; Jierry, L.; Bouerat, L.; Schmitt, M.
Tetrahedron 2002, 58, 4195.
´
17. For assignment of the (+)-(R,R) configuration of the
stilbene oxide, see: Imuta, M.; Ziffer, H. J.Org.Chem.
1979, 44, 2505; For the assignment of (+)-(2S,3R)-config-
uration to the methyl p-methoxycinnamate epoxide, see:
Matsuki, K.; Sobukawa, M.; Kawai, A.; Inoue, H.;
Takeda, M. Chem.Pharm.Bull. 1993, 41, 643 (+)-(R,R)-
and (ꢀ)-(S,S)-b-methylstyrene oxide are commercially
available: cf. Aldrich Chiral catalogue.
References
1. Mello, R.; Fiorentino, M.; Fusco, C.; Curci, C. J.Am.
Chem.Soc. 1989, 111, 6749.
2. Denmark, S. E.; Wu, Z.; Crudden, C. M.; Matsuhashi, H.
J. J.Org.Chem. 1997, 62, 8288.
18. Armstrong, A.; Washington, I.; Houk, K. N. J.Am.
Chem.Soc. 2000, 122, 6297.
19. Baumstark, A. L.; McKloskey, C. J. Tetrahedron Lett.
1987, 28, 3311.
20. Yang, D.; Wang, X. C.; Wong, M. K.; Yip, Y. C.; Tang,
M. W. J.Am.Chem.Soc. 1996, 118, 11311.
21. Adam, W.; Paredes, R.; Smerz, A. K.; Veloza, L. A.
Liebigs Ann. 1997, 547.
22. Jenson, C.; Liu, J.; Houk, K. N.; Jorgensen, W. L. J.Am.
Chem.Soc. 1997, 119, 12982.
23. Direct epoxidation by oxone provided some of the racemic
epoxide, which thus lowered the enantioselectivity
observed. However there is a direct epoxidation but with
b-methylstyrene.
3. (a) Armstrong, A.; Hayter, B. R. Chem.Commun. 1998,
621–622; (b) Tu, Y.; Wang, Z. X.; Frohn, M.; He, M.; Yu,
H.; Tang, Y.; Shi, Y. J.Org.Chem. 1998, 63, 8475; (c)
Yang, D.; Yip, Y. C.; Chen; Cheung, K. K. J.Am.Chem.
Soc. 1998, 120, 7659–7660.
4. Denmark, S. E.; Wu, Z. Synlett 1999, 847.
´ ´
5. Solladie-Cavallo, A.; Bouerat, L. Org.Lett. 2000, 2, 3531–
3534.
6. Solladie-Cavallo, A.; Jierry, L.; Norrousi-Arasi, H.; Tah-
´
massebi, D. J.Fluorine Chem. 2004, 125, 1371–1377.
7. Armstrong, A.; Moss, W. O.; Reeves, J. R. Tetrahedron:
Asymmetry 2001, 12, 2779.
8. (a) Armstrong, A.; Ahmed, G.; Dominguez-Fernandez,
B.; Hayter, B. R.; Wailes, J. S. J.Org.Chem. 2002, 67,