Copper-catalyzed asymmetric ring-opening reaction of oxabenzonorbornadienes with Grignard and aluminum reagents

Article abstract of DOI:10.1055/s-0029-1216838

A highly enantioselective method for the copper-catalyzed desymmetrization of oxabenzonorbornadienes with aluminum reagents and SimplePhos as chiral ligand has been developed. The same reaction with Grignard reagents is also reported. A wide range of alky

Full text of DOI:10.1055/s-0029-1216838

SPECIAL TOPIC
2101
Copper-Catalyzed Asymmetric Ring-Opening Reaction of
Oxabenzonorbornadienes with Grignard and Aluminum Reagents
A
symmetric Ring
e
-Opening
n
eaction of
O
a
xabenzono
u
rbornadiene
d
Millet, Ludovic Gremaud, Tania Bernardez, Laëtitia Palais, Alexandre Alexakis*
Department of Organic Chemistry, University of Geneva, 30 Quai Ernest Ansermet, 1211 Geneva 4, Switzerland
Fax +41(22)3793215; E-mail: alexandre.alexakis@chiorg.unige.ch
Received 7 April 2009
form the anti-adduct with organometallic reagents.
Feringa and co-workers applied a chiral copper–phos-
phoramidite catalyst to the addition of dialkylzinc to oxa-
benzonorbornadienes.10 Following this report, Grignard
Abstract: A highly enantioselective method for the copper-cata-
lyzed desymmetrization of oxabenzonorbornadienes with alumi-
num reagents and SimplePhos as chiral ligand has been developed.
The same reaction with Grignard reagents is also reported. A wide
range of alkyl chains have been used with moderate to high enantio- reagents were applied with copper catalysis in non-enan-
tioselective methods by Carretero and co-workers.11 the
selectivity and high trans selectivity. The transfer of a methyl group
is also reported with high enantiomeric and diastereomeric excess
enantioselective version was described by Zhou and co-
and yields for different substrates. We finally have been able to per-
workers.12 The anti stereoselectivity of the reaction comes
form the first trans enantioselective desymmetrization of oxaben-
from a pure SN2¢ mechanism.10,11a However limitations
exist in all cases. Despite high yields and enantioselectiv-
zonorbornadienes with an aromatic group.
Key words: copper catalysis, asymmetric transformation, Grignard
ities obtained with dialkylzinc reagents, transfer of the
reagents, aluminum reagents, SimplePhos
methyl group occurred with low yield and conversion.
Furthermore, one equivalent of zinc triflate is needed for
the reaction to proceed.10 Grignard reagents are far more
Formation of multiple stereocenters is one of the most im-
reactive, but the enantioselective version suffers high lim-
portant issues in modern organic chemistry. The desym-
itations as only primary and linear alkyl Grignard reagents
metrization of meso compounds has attracted much
can be transferred.12 Another lack in both methodologies
is the absence of transfer of aromatic groups. Transfer of
attention since many stereocenters can be easily estab-
lished from simple transformations.1 During the last de-
aryl groups in an anti fashion is only observed in racemic
cade, desymmetrization of oxabicyclic alkenes with
versions of the reaction.11
transition metal catalysis has exploded.2 Most of the de-
Recently, our group reported the synthesis of a new family
scribed methodologies provide the formation of the syn-
adduct.
of ligands: SimplePhos (e.g., L1, Figure 1).13
A large variety of carbon nucleophiles have been used to
perform this transformation. Initially organolithium
reagents2a,3 and organocuprates4 showed limited success
in affording syn-addition products in moderate to good
P
N
P
N
yields. Only the introduction of softer organometallic nu-
cleophiles in combination with transition metal catalysis
allowed the asymmetric desymmetrization of oxabicyclic
alkenes in good yields. Among all the nucleophiles used
for this transformation, organozinc reagents in combina-
tion with palladium catalysis have been the most studied.5
Other nucleophiles such as aluminum reagents in combi-
nation with titanium or zirconium,6 organoboron with
rhodium7 or palladium8 and Grignard reagents with
zirconium6 have also been employed. All these methodol-
ogies provide the syn-adduct resulting from an attack on
the exo face of the oxabicyclic unit.
L1
Figure 1 SimplePhos ligand (L1)
This family of ligands showed high efficiency in copper-
catalyzed allylic alkylation and in desymmetrization of bi-
cyclic hydrazines with trialkylaluminum reagents13 as
well as in SN2¢ ring-opening reactions of vinyloxiranes.14
In the course of our studies, we decided to investigate the
application of SimplePhos ligands in copper-catalyzed
desymmetrization of oxabenzonorbornadienes with
Grignard and aluminum reagents.
The anti-ring-opening reaction is less documented. The
first report was from Lautens and co-workers with the
rhodium-catalyzed desymmetrization of oxabenzonorbor-
nadienes with heteroatomic nucleophiles.9 Copper-cata-
lyzed allylic alkylation appeared to be an efficient tool to
Grignard Reagents
To complete our study on the copper-catalyzed allylic
alkylation with Grignard reagents,15 we turned our atten-
SYNTHESIS 2009, No. 12, pp 2101–2112
Advanced online publication: 19.05.2009
DOI: 10.1055/s-0029-1216838; Art ID: C01409SS
© Georg Thieme Verlag Stuttgart · New York
x
x.
x
x
.
2
0
0
9

2102
R. Millet et al.
SPECIAL TOPIC
OH
O
OH
EtMgBr (1.2 equiv)
CuTC (3 mol%)
Et
Et
+
L* (3 mol%)
CH2Cl2, –20 °C, 20 h
1
anti-2a
syn-2a
Scheme 1 General procedure for the determination of the family of ligands
Zhou12 explained that only nonpolar solvents gave good
conversions and CuTC is known to be efficient in dichlo-
romethane.
Ph
N
Ph
N
O
O
O
O
P
P
The phosphoramidite family confirms the observation
made by Zhou and co-workers that only spiro-phosphor-
amidites gave interesting enantiomeric excesses.12a The
atropos ligand L2 showed an important match/mismatch
effect as the match (S,S,S)-L2 gave 2a in good conversion,
high diastereoselectivity but moderate enantiomeric ex-
cess (Table 1, entry 1); whereas the mismatch diastere-
omer (R,S,S)-L2 gave only 43% conversion with
moderate diastereoselectivity and an even lower enantio-
meric excess (Table 1, entry 2). These two results show
that the amine part-controls the enantioselectivity of the
reaction. The tropos L3 was slightly below (S,S,S)-L2 in
terms of enantioselectivity (Table 1, entry 3). The Josi-
Phos ligand L4 appears efficient in term of conversion
and diastereoselectivity but only 10% ee was obtained
(Table 1, entry 4). Carbene ligand L516 was also tested,
but the reaction was very slow and low selectivities were
obtained (Table 1, entry 5). We then turned our attention
to the SimplePhos ligand L1.13 We were delighted to see
that in terms of conversion (97% conversion), diastereo-
selectivity (ratio anti/syn 98:2) and enantioselectivity
(82% ee) good results were obtained (Table 1, entry 6).
The absolute configuration of anti-2a was determined to
Ph
Ph
(S,S,S)-L2
(R,S,S)-L2
Ph
N
O
O
P
Ph
L3
Cl–
P(C6H11 2
)
Ph2P
t-Bu
Fe
N
N
HO
L4
L5
Figure 2 Preliminary tested ligands
Table 1 Preliminary Test of Ligands for the Desymmetrization of
Oxabenzonorbornadiene 1 (Scheme 1)
Entry
Ligand
Product 2a
Conv.a (%) Ratiob anti/syn eeb (%) anti
20
be (1S,2R) by comparison with literature data {[a]D
–264.2 (c 1.01) 82% ee; 1S,2R)}.10
1
2
3
4
5
6
(S,S,S)-L2
87
43
98:2
90:10
92:8
98:2
62:38
98:2
56 (–)
23 (–)
36 (–)
10 (+)
13 (–)
82 (–)
Once a good family of ligands was determined, we inves-
tigated the best conditions for this reaction. First of all the
influence of the copper salt and the solvent were studied
as their influence is known to be important in copper-
catalyzed allylic alkylations (Table 2).15,17,18
(R,S,S)-L2
L3
L4
L5
L1
85
100
10
First of all the effect of the solvent was investigated. It
proved to have a dramatic effect on the enantioselectivity
of the reaction. When the reaction was carried out in
97
a Determined by GC-MS and 1H NMR.
dichloromethane,
2-ethyl-1,2-dihydronaphthalen-1-ol
b Determined by chiral GC, the sign of the optical rotation is in paren-
theses.
(2a) was isolated in 90% yield with an anti/syn ratio of
98:2 and 82% ee (Table 2, entry 1). The solvent was then
changed from dichloromethane to toluene; a dramatic ef-
fect was observed on the enantioselectivity as it dropped
to 47% (Table 2, entry 2). As expected,12a diethyl ether
was an even worse solvent as the alcohol 2a was formed
in nearly racemic form and a large amount of naphthalene
was observed (Table 2, entry 3). To confirm the tendency
with toluene, copper(II) triflate was used as the copper salt
to eliminate the possibility of a unfavorable combination
between toluene and CuTC. As no improvement was ob-
served by changing the copper salt, we chose dichlo-
romethane as the solvent (Table 2, entry 4).
tion to the desymmetrization of oxabenzonorbornadienes
(Scheme 1).
We began our studies with the addition of ethylmagne-
sium bromide to oxabenzonorbornadiene 1. First of all,
the classical families of ligands (Figure 2) were tested to
determine which family would be the more efficient (Ta-
ble 1).
The tests were carried out in dichloromethane with copper
thiophene-2-carboxylate (CuTC) as copper salt. The
choice of the solvent was easy as both Carretero11 and
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

SPECIAL TOPIC
Asymmetric Ring-Opening Reaction of Oxabenzonorbornadienes
2103
Table 2 Copper Salt and Solvent Screeninga
Table 3 Study of the Temperature
OH
OH
O
OH
OH
O
EtMgBr (1.2 equiv)
CuX (3 mol%)
Et
Et
Et
Et
+
EtMgBr (1.2 equiv)
CuTC (3 mol%)
+
L1 (3 mol%)
solvent, –20 °C, 20 h
L1 (3 mol%)
CH2Cl2, T °C, 20 h
1
anti-2a
syn-2a
anti-2a
syn-2a
1
Entry Solvent
Cu salt
Conv.b
Ratioc
eec (%)
Entry
Temp (°C) Conv.a (%) Ratiob anti/syn eeb (%) anti
(%)
anti/syn anti
1
2
3
–20
22
97 (90)
70c
98:2
80:20
98:2
82
56
78
1
2
CH2Cl2
toluene
Et2O
CuTC
97 (90) 98:2
82
47
8
CuTC
98
88
97:3
95:5
95:5
98:2
98:2
95:5
98:2
98:2
99:1
–40
90
3
CuTC
a Determined by GC-MS and 1H NMR, isolated yield is in parenthe-
ses.
4
toluene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Cu(OTf)2
Cu(OTf)2
CuBr
98
50
78
80
76
75
84
80
b Determined by chiral GC.
c 13% of 1,2-dihydronaphthalen-1-ol detected.
5
93
6
98
influence of each part of the ligand, we envisaged modi-
fying the bulk around the phosphorus part.
7
CuCl
61
The steric bulk was increased in ligand L6 and decreased
in ligand L7.19 When using L6 as ligand, a decrease in the
reactivity as well as in the enantioselectivity was observed
(Table 4, entry 2).
8
Cu(OAc)2·H2O
CuTC
98
9e
10f
100
91
CuTC
Going from the two phenyl groups to two methyl groups
had a dramatic effect on the enantioselectivity of the reac-
tion as 2a was obtained with only 28% ee (Table 4, entry
3). The attempt to decrease the rotation rate between the
two aryl groups was also not beneficial to the enantiose-
lectivity (Table 4, entry 4). The bulk of the amino part was
increased with L9, but the result obtained was almost the
same than with L1 (Table 4, compare entries 1 and 5). Fi-
nally, adding chelating groups on the aromatic groups of
the amino part was detrimental to the reaction as the con-
version dropped to less than 10% after 20 hours and the
enantiomeric excess was only 60% (Table 4, entry 6).
a Reaction conditions: 1 (1 mmol), Cu (3 mol%), L1 (3 mol%), EtMg-
Br in Et2O(1.2 equiv), solvent (6 mL); order of addition: substrate
then Grignard reagent.
b Determined by GC-MS and 1H NMR, the isolated yield is in paren-
theses.
c Determined by chiral GC, absolute configuration (–)-(1S,2R).
d Naphthalene (25%).
e L1 (6 mol%).
f Reverse addition order: Grignard reagent was added, then the sub-
strate.
The effect of the copper salt is less important. Different
copper(I) or copper(II) salts were tested (Table 2, entries
5–8). Only with copper(I) chloride was a decrease of the
conversion observed (Table 1, entry 7). However CuTC
remains slightly above the others copper salts. To com-
plete the study the ratio ligand/copper was increased to
2:1. The enantioselectivity slightly increased from 82% to
84% ee, but 2-ethylnaphthalene was observed as a
byproduct (Table 2, entry 9). Finally the reverse addition
order was tested without any improvement (Table 2, entry
10).
AlMe3
Ph
N
Ph
N
Ph
N
Me
Me
P
P
P
Ph
Ph
Ph
The temperature was then changed. When the reaction
was performed at room temperature, the enantioselectivi-
ty decreased to 56% ee and the anti/syn ratio decreased to
80:20 (Table 3, entry 2). It should be noted that a large of
amount of reductive ring opening leading to racemic 1,2-
dihydronaphthalen-1-ol occurred. This may be due to a b-
hydride elimination on the cuprate, forming a copper hy-
dride species. In order to increase the enantioselectivity of
the reaction, the temperature was decreased to –40 °C, but
no improvement was observed (Table 3, entry 3).
L8
L6
L7
OMe
OMe
Ph
Ph
Ph
Ph
P
N
P
N
The ligand was then modified to study its influence on the
outcome of the reaction (Figure 3). To have an idea of the
L9
L10
Figure 3 Ligands used for this study
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

2104
R. Millet et al.
SPECIAL TOPIC
Table 4 Modification of SimplePhos Ligand for the Desymmetriza-
tion of Oxabenzonorbornadiene 1 (Scheme 1)a
Table 5 Scope of the Grignard Reagenta
OH
R
O
RMgBr (1.5 equiv)
CuTC/L1 (3 mol%)
Entry
Ligand
Product 2a
Conv.b (%) Ratioc anti/syn eec (%) anti
CH2Cl2, –20 °C, 20 h
1
2a–h
1
2
3
4
5
6
L1
L6
L7
L8
L9
L10
93 (90)
65
98:2
92:8
99:1
96:4
99:1
92:8
82
62
28
74
80
60
Entry
R
Product Yieldb
(%)
Ratioc
eec (%)
anti/syn anti
100
1
Et
2a
2b
2b
2c
2d
2e
90
(<10)
84
98:2
–
82
71
72
76
62
57
93
98
9
2
Me
3d
4
Me
90:10
98:2
95:5
98.2
n-Bu
i-Bu
(CH2)2Ph
86
a Reaction conditions: 1 (1 mmol), CuTC (3 mol%), L (3 mol%), Et-
MgBr in Et2O (1.2 equiv), CH2Cl2 (6 mL), –20 °C; order of addition:
substrate then Grignard reagent.
5
(73)
64
b Determined by GC-MS and 1H NMR in parentheses isolated yield.
c Determined by chiral GC.
6
7
2f
95
97:3
64
Before broadening the scope of the reaction, the reaction
was performed with ethylmagnesium chloride to check if
the same halogen effect exists with SimplePhos ligand as
with phosphoramidite ligands (Scheme 2).12a
8
9f
i-Pr
i-Pr
Ph
2g
2g
2h
75e
(65)
89
90:10
95:5
98:2
30
54
0
10
a Reaction conditions: 1 (1 mmol), CuTC (3 mol%), L1 (3 mol%),
RMgBr in Et2O(1.2 equiv), CH2Cl2 (6 mL), –20 °C; order of addition:
substrate then Grignard reagent.
OH
O
EtMgCl (1.5 equiv)
Et
85% conversion
67% of 2a
anti/syn 97:3
36% ee
CuTC/L25 (3 mol%)
CH2Cl2, –20 °C, 20 h
b Isolated yield, conversion determined by GC-MS and 1H NMR in
parentheses.
anti-2a
1
c Determined by chiral GC.
d Reaction run at r.t.
Scheme 2 Effect of the Grignard reagent counterion
e Naphthalene (20%) was detected by 1H NMR.
f Reaction run at –40 °C.
The conversion after 20 hours was 85% [2a (67%), naph-
thalene (17%), and 2-ethylnaphthalene (1%)]. The enan-
tiomeric excess dropped to only 36%. The role of the
counterion of the Grignard reagent appears to be crucial to
the enantioselectivity of the reaction. However, it does not
seem to play a role in the catalytic cycle as copper(I) chlo-
ride gave the same level of enantioselectivity as CuTC
with ethylmagnesium bromide (Table 2, entries 1 and 7).
ee (Table 5, entry 5). Phenethylmagnesium bromide in-
duced lower enantioselectivity in 2e (Table 5, entry 6).
The functionalized 4-methylpent-3-enylmagnesium bro-
mide gave result in the same range, trans-2-(4-methyl-
pent-3-enyl)-1,2-dihydronaphthalen-1-ol
(2f)
was
obtained in 95% yield, with high anti/syn ratio but moder-
ate 64% ee (Table 5, entry 7). Moving to secondary
Grignard reagents was more problematic, at –20 °C iso-
propylmagnesium bromide gave full conversion, however
naphthalene was also formed. The alcohol 2g was ob-
tained in moderate anti/syn ratio (90:10) and only 30% ee
(Table 5, entry 8). Decreasing the temperature to –40 °C,
decreased the conversion but also the amount of side
product (10% of naphthalene). An increase in the anti/syn
ratio as well as the enantiomeric excess was also observed
(Table 5, entry 9). Finally phenylmagnesium bromide was
tested and the corresponding alcohol 2h was obtained in
89% yield with high diastereoselectivity but as a racemic
mixture (Table 5, entry 10).12 Finally, we investigated the
substrate scope for the reaction (Figure 4).
With the set of conditions in hand, the scope of the
Grignard reagent was then studied (Table 5). Addition of
ethylmagnesium bromide gave trans-2-ethyl-1,2-dihy-
dronaphthalen-1-ol (anti-2a) in 90% yield and 82% ee
(Table 5, entry 1).
Methylmagnesium bromide appeared to be less reactive
than ethylmagnesium bromide. The temperature had to be
increased to room temperature in order to have full con-
version (Table 5, entries 2 and 3). 2-Methyl-1,2-dihy-
dronaphthalen-1-ol (2b) was obtained in 84% yield with
ratio anti/syn 90:10 and 72% ee (Table 5, entry 3); at this
temperature naphthalene was detected as a side product.
When a longer alkyl chain was used the enantioselectivity
remains in the same range (Table 5, entry 4). The bulk of
the alkyl chain was increased in the b-position going to an
isobutyl chain. Despite good conversion and diastereose-
lectivity, the enantioselectivity for 2d decreased to 62%
We wanted to modify the aromatic ring of the oxaben-
zonorbornadiene (Table 6) and also to apply the method-
ology to nonbenzylic oxabicyclic alkenes.
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

SPECIAL TOPIC
Asymmetric Ring-Opening Reaction of Oxabenzonorbornadienes
2105
substrate bearing the two methoxy groups in the ortho po-
sitions could not be desymmetrized under these condi-
tions; only byproducts such 1,4-dimethoxynaphthalene
and 6-ethyl-1,4-dimethoxynaphthalene were formed
(Table 6, entry 5).
OMe
F
O
F
MeO
MeO
O
O
OMe
3
4
5
OH
EtMgBr (1.2 equiv)
R1
OMe
OMe
CuTC/L1 (3 mol%)
Et
O
OH
OH
OMe
OMe
CH2Cl2, –20 °C, 20 h
O
O
6
12
R2
60% conversion
anti/syn 90:10
0% ee
6
7, R1 = H, R2 = Me
8, R1 = R2 = Me
Figure 4 Substrates studied
Table 6 Substrate Scope; Oxabenzonorbornadienes Derivatives
Scheme 3 Desymmetrization of a nonbenzylic substrate
When nonbenzylic substrate 6 was used the result was not
enthusiastic (Scheme 3). The reaction was clean even if
the conversion was not complete after 20 hours; the result-
ing alcohol 12 was obtained in racemic mixture in a ratio
anti/syn 90:10. However this substrate is known to be very
challenging as its asymmetric desymmetrization in anti
fashion has not previously been reported.
R2
OH
R2
R3MgBr (1.5 equiv)
CuTC/L1 (3 mol%)
R1
R1
R3
R1
R1
O
CH2Cl2, –20 °C, 20 h
R2
R2
3, R1 = F, R2 = H
4, R1 = OMe, R2 = H
5, R1 = H, R2 = OMe
9a, 10a,b, 11b
The described methodology with Grignard reagents
showed interesting results with the simple oxabenzonor-
bornadiene 1 compared to the previous published results
with Grignard reagents,12 but is very limited in substrate
scope. To overcome this problem, we turned our attention
to the aluminum reagents.
Entry Substrate R3
Product Conv.a
(%)
Ratiob
anti/syn
eec
(%)
1
3
4
4
4
5
Et
9a
10b
10a
10a
11b
100 (85)
40e
96:4
64
20
20
31
–
2d
3
Me
Et
>95:5
Aluminum Reagents
100 (53)f >95:5
The major problem with dialkylzinc reagents and oxaben-
zonorbornadiene is their low reactivity. To obtain good
conversion Feringa and co-workers had to add one equiv-
alent of zinc triflate to act as a Lewis acid.10 Organoalumi-
num reagents are known to be more Lewis acidic than
their zinc analogues. This property combined with the
high oxophilicity of aluminum (Al–O bond strength 138
kcal/mol) made aluminum reagents interesting candidates
to act as nucleophiles in the desymmetrization reaction of
oxabenzonorbornadiene derivatives. Moreover as they are
produced on industrial scale, they are inexpensive and a
large variety of primary trialkylaluminum are commer-
cially available (Me3Al, Et3Al, n-Pr3Al, n-Bu3Al, i-
Bu3Al). Vinylaluminums and arylaluminums are also eas-
ily available via hydroalumination of alkynes20 or Li/Al
exchange.21
4g
5
Et
50h
>95:5
–
i
Me
–
a Determined by GC-MS and 1H NMR, isolated yield is in parenthe-
ses.
b Determined by 1H NMR.
c Determined by chiral SFC.
d Reaction run at r.t.
e 2,3-Dimethoxynaphthalene (11%) was detected.
f 2,3-Dimethoxynaphthalene (37%) was detected.
g Reaction run at – 40 °C.
h 2,3-Dimethoxynaphthalene (25%) was detected.
i No traces of the desired product.
The electron density of the aromatic group plays a very
important role in the control of the reaction. Decreasing
the electron density of the aromatic ring induced a de-
crease in the enantiomeric excess to 64% (Table 6, entry
1). Electron-rich substrates were then studied. Substrate 4
reacted with both methyl- and ethylmagnesium bromide
to give the corresponding alcohols 10b and 10a in low
yield and with very low enantiomeric excess (Table 6, en-
tries 2 and 3). A large amount of 2,3-dimethoxynaphtha-
lene was also formed. Decreasing the temperature to –40
°C with ethylmagnesium bromide allowed a slight in-
crease in the enantioselectivity to 31% ee, but still a large
amount of byproduct was observed (Table 6, entry 4). The
Based on our study with Grignard reagents, SimplePhos
ligands were chosen as ligands to determine the best con-
ditions. The copper salt and solvent screening was then
performed (Table 7).
The counterion of the copper salt has a dramatic effect on
the outcome of the reaction. CuTC leads to a very clean
reaction with excellent diastereoselectivity and 89% ee
(Table 7, entry 1). When OTf was used as the counterion,
despite complete conversion no trace of 2b was detected,
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

2106
R. Millet et al.
SPECIAL TOPIC
only 1-naphthol and 1-methylnaphthalene were detected, conversion without lost of enantioselectivity. However a
whatever the oxidation state of the copper (Table 7, en- small decrease in diastereoselectivity was observed (Ta-
tries 2 and 3). When bromine was used as the counterion, ble 7, entry 12). Finally, methyl tert-butyl ether (MTBE)
the product was obtained in moderate conversion, diaste- appeared to be the solvent of choice with full conversion
reoselectivity and enantioselectivity (35%, 80:20, 50% ee, within 20 hours in 93% isolated yield in perfect diastereo-
respectively) in both cases (Table 7, entries 5 and 6). In selectivity and high enantiomeric excess (Table 7, entry
the same family of counterion, copper(I) chloride gave in- 13). The temperature was decreased to 0 °C but no real
teresting results with 81% ee for 40% conversion and ratio improvement was observed (Table 7, entry 14). In order to
anti/syn 90:10 (Table 7, entry 4), whereas copper(I) iodide improve the enantiomeric excess, modification of the
gave no reaction (Table 7, entry 7). Copper(I) acetate ligand was envisaged.
monohydrate showed low reactivity but high selectivities
(Table 7, entry 8). As the counterion seems to be influen-
tial on the outcome of the reaction, we decided to try the
noncoordinated tetrakis(acetonitrile)copper(I) tetrafluo-
roborate as the copper salt (Table 7, entry 9). Despite good
diastereo- and enantioselectivities (95:5, 88% ee), a large
amount of byproduct was formed (Table 7, entry 9). It ap-
Modifications of the SimplePhos ligand were studied first.
An increase of the bulk on the phosphorus part of L6 did
not lead to any improvement. The resulting anti-product
anti-2b was obtained as a single diastereoisomer in 90%
ee (Table 8, entry 2). The amino part of the ligand was
modified from a diphenyl to a bis(2-naphthyl) in L9, but
the enantiomeric excess did not increase significantly (Ta-
peared that CuTC is the copper salt of choice for such a re-
ble 8, entry 3). It appears that neither modification of the
action. Then different solvents were tested; the conversion
phosphorus part nor of the amino part influenced the out-
decreased to only 15% when toluene was used as solvent
come of the reaction. At this stage we turned our attention
(Table 7, entry 10) and no reaction was observed in tet-
to phosphoramidite ligand as the reaction was performed
rahydrofuran (Table 7, entry 11). The conversion issue
in coordinating solvent.17 (S,S,S)-L2 led to the formation
was solved by using diethyl ether as solvent with 90%
Table 7 Copper Salt and Solvent Screening
OH
OH
Me3Al (1.5 equiv)
CuX (3 mol%)
+
O
L1 (3 mol%)
solvent, r.t., 20 h
1
anti-2b
syn-2b
Entry
Copper
CuTC
Solvent
Conv.a (%)
Ratiob anti/syn
eeb (%)
89
–
1
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
toluene
THF
54
–
99:1
–
2c
3c
4
Cu(OTf)2
CuOTf·C6H6
–
–
–
CuCl
40d
35
90:10
80:20
80:20
–
81
50
50
–
5
CuBr
6
CuBr·Me2S
CuI
36
7
0
8
Cu(OAc)·H2O
Cu(MeCN)4·BF4
CuTC
30
95:5
95:5
–
89
88
–
9
66e
15
10
11
12
13
14f
CuTC
0
–
–
CuTC
Et2O
90
91:9
99:1
99:1
90
91
92
CuTC
MTBE
MTBE
100 (93)
100
CuTC
a Determined by 1H NMR, isolated yield is in parentheses.
b Determined by chiral GC.
c Only 1-naphthol (40%) and 1-methylnaphthalene (60%) were detected.
d 1-Naphthol and 1-methylnaphthalene were detected.
e 1-Naphthol (20%) was detected.
f Reaction performed at 0 °C
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

SPECIAL TOPIC
Asymmetric Ring-Opening Reaction of Oxabenzonorbornadienes
2107
of 2b with 66% ee with a change in the selectivity (Table
8, entry 4), whereas (R,S,S)-L2 allowed the formation of
2b with only 45% ee (Table 8, entry 5). The biphenyl-
based phosphoramidite L3 gave a less selective reaction
(Table 8, entry 6). As the SimplePhos ligand L1 gave the
most elective reaction, it was used for further investiga-
tions. To extend the scope of the reaction, other aluminum
reagents and oxabenzonorbornadiene derivatives 3, 4, and
5 were investigated.
Table 8 Influence of the Ligand
OH
OH
Me3Al (1.5 equiv)
CuTC (3 mol%)
+
O
L* (3 mol%)
MTBE, r.t. 20 h
1
anti-2b
syn-2b
Entry
Ligand
L1
Conv.a (%) Ratioa anti/syn eeb (%)
1
2
3
4
5
6
100
100
100
100
100
100
99:1
99:1
99:1
99:1
99:1
99:1
91 (–)
90 (–)
92 (–)
66 (+)
45 (–)
57 (–)
First of all, the enantioselective reduction of 1 with di-
isobutylaluminum hydride was envisaged. Unfortunately
under the reaction conditions only naphthalene was
formed (Table 9, entry 1). All the commercially available
trialkylaluminum reagents were tested on oxabenzonor-
bornadiene 1. To our great delight only a slight decrease
in the yield was observed going from trimethyl- to trieth-
ylaluminum and the same range of enantiomeric excesses
were observed (Table 9, entries 2 and 3). Even longer
alkyl can be introduced without loss of enantioselectivity
(Table 9, entries 4 and 5). The low isolated yield with
tripropylaluminum is probably due to a concentration
problem of the aluminum in the bottle as tributylalumi-
num gave high isolated yield. We were very surprised to
L6
L9
(S,S,S)-L2
(R,S,S)-L2
L3
a Determined by 1H NMR.
b Determined by chiral GC, sign of optical rotation is in parentheses.
Table 9 Substrate and Aluminum
2j, R1 = R2 = H, R = H
R2
OH
R2
2a, R1 = R2 = H, R = Et
R33Al (1.5 equiv)
CuTC/L1 (3 mol%)
R1
R1
R3
R1
R1
2i, R1 = R2 = H, R = n-Pr
2c, R1 = R2 = H, R = n-Bu
2d, R1 = R2 = H, R = i-Bu
7b, R1 = F, R2 = H, R = Me
8a, R1 = OMe, R2 = H, R = Et
8b, R1 = OMe, R2 = H, R = Me
9a, R1 = H, R2 = OMe, R = Et
9b, R1 = H, R2 = OMe, R = Me
9d, R1 = H, R2 = OMe, R = i-Bu
O
MTBE, r.t., 20 h
R2
R2
1, R1 = R2 = H
2a,c,d,i,j, 9b, 10a,b, 11a–c
3, R1 = F, R2 = H
4, R1 = OMe, R2 = H
5, R1 = H, R2 = OMe
Entry
1d
2
Substrate
R3
Product
2j
Yielda (%)
Ratiob anti/syn
eec (%)
–
1
1
1
1
1
1
3
4
4
5
5
5
H
(100)e
93
–
Me
Et
2b
2a
99:1
99:1
99:1
99:1
99:1
94:6
99:1
99:1
93:7
99:1
–
94
92
87
87
94
91
88
89
87
88
–
3
79
4
n-Pr
n-Bu
i-Bu
Me
Me
Et
2i
50
5
2c
95
6
2d
71
7
9b
85
8
10b
10a
11b
11a
11d
69
9
95
10
11
12
Me
Et
88
(100)
–
i-Bu
a Isolated yield, the conversion is in parentheses.
b Determined by 1H NMR.
c Determined by chiral GC or SFC.
d DIBAL-H was used as nucleophile.
e No traces of the desired alcohol were detected.
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

2108
R. Millet et al.
SPECIAL TOPIC
see that even triisobutylaluminum was transferred in good no improvement was found on the relative transfer of phe-
yield and with a high enantioselectivity (94% ee) (Table nyl vs alkyl group, but the enantiomeric excess increased
9, entry 6). The electron-deficient oxabenzonorborna- to 84% (Scheme 5).
diene derivative 3, reacted with trimethylaluminum to
give 9b in good yield and 91% ee (Table 9, entry 7). High
enantioselectivities were also achieved with electron-en-
riched substrates (Table 9, entries 8–11). A large amount
Finally, we have applied the previously described meth-
odology to substituted oxabenzonorbornadienes, but in
the non-aromatic ring (Table 10). In this series, an unex-
pected byproduct 14 was observed, probably due to a cat-
of 2,3-dimethoxynaphthalene was obtained when 4 was
ionic rearrangement.
reacted with trimethylaluminum (Table 9, entry 8),
The monosubstituted oxabenzonorbornadiene 7 gave
whereas using triethylaluminum gave 10a in high yield
moderate to good result with different organoaluminum
and enantioselectivity (Table 9, entry 9). When trimethyl-
reagents as nucleophiles (Me3Al, Et3Al, n-Pr3Al, Ph3Al).
The reaction with 7 (which is chiral, but racemic) corre-
sponds to a kinetic resolution. Therefore, only 0.5 equiv-
alent of R3Al was added, and the maximum expected yield
aluminum was used on 5 the corresponding alcohol 11b
was obtained in good yield (Table 9, entry 10). However,
the ethyl adduct 11a could not be purified due to aromati-
zation. Finally triisobutylaluminum was tested on 5, but
should be 50%. The Lewis character of the aluminum
no traces of the desired alcohol were detected (Table 9,
atom plays a very important role and induces the ring
opening of the oxygen bridge to form two secondary prod-
ucts 14 and 15. If this Lewis character increases, more
byproducts are produced, and in this case the diastereo-
meric excesses and enantiomeric excesses increase (Table
entry 12). We next turn our attention to the nonbenzylic
substrate 6 (Scheme 4). We were disappointed to see that
no reaction occurred.
OH
Et3Al (1.2 equiv)
10, entries 1–4). The anti/syn ratios obtained for sub-
strates 7 and 8 are very good (99:1) (Table 10, entries 1–
6) and in all cases in favor of the anti-product. The disub-
stituted oxabenzonorbornadiene 8 gave moderate conver-
sion 75–77% and very good enantiomeric excess 94–97%
(Table 10, entries 5 and 6). However the formation of
product 13e was dramatically dependent on the copper
salt. If copper(I) chloride is replaced by copper(I) acetate
monohydrate, the expected product 13e is the major one.
CuTC/L1 (3 mol%)
OMe
OMe
Et
O
OH
OH
MTBE, r.t., 20 h
6
10b
0% conversion
Scheme 4 Attempt of desymmetrization of 6 with triethylaluminum
We decided to take advantage of the easy formation of vi-
nyl-20 and arylaluminum reagents21 to have access to new
compounds. The tandem hydroalumination/asymmetric
allylic alkylation was performed using the standard condi-
tions (3 mol% of catalyst) but no reaction was observed.
We then increased the catalyst loading to 10 mol%, but
only the reduced ring-opening product was observed in a
racemic mixture. The tandem lithium–aluminum ex-
change/asymmetric allylic alkylation was then performed
(Scheme 5).
In summary, we have shown that SimplePhos ligands are
efficient ligands to perform the desymmetrization of oxa-
benzonorbornadiene 1 with Grignard reagents. However
this reaction is substrate limited as modification of the ar-
omatic cycle or the use of nonbenzylic substrate is very
detrimental to the reaction. However we were able to
transfer bulky groups such as isobutyl or phenethyl that
were not transferable under the previously described con-
ditions with Grignard reagents.12a The aluminum reagents
are far more interesting as the only limitation is with sub-
strate 6, which is still challenging in copper-catalyzed de-
symmetrization. The aluminum reagents are Lewis acidic
enough to react without additives in high yield and enan-
tioselectivity. Finally, we have described the first copper-
catalyzed asymmetric desymmetrization of oxabenzonor-
bornadienes with aryl groups with high enantioselectivity.
Three reactants were used to transmetalate from lithium to
aluminum. When dimethylaluminum chloride was used,
the phenyl group was transferred in 42% with 71% ee,
whereas the methyl group was transferred in 58% with
90% ee. To increase the amount of phenyl group transfer
we envisaged using diethylaluminum chloride or diisobu-
tylaluminum chloride as transmetalating agents. However
R2AlCl (3 equiv)
PhLi (3 equiv)
PhAlR2 (3 equiv)
–30 °C, 30–60 min
solvent
OH
OH
Ph
R
CuTC/L1 (3 mol%)
O
+
MTBE, r.t., 20 h
1
2h
2b, R = Me
2a, R = Et
2d, R = i-Bu
R = Me, 2h/2b 42:58, 2h 71% ee, 2b 90% ee
R = Et, 2h/2a 40:60, 2h 84% ee, 2a 94% ee
R = i-Bu, 2h/2d 35:65, 2h 84% ee, 2d 94% ee
Scheme 5 Tandem Li–Al exchange/asymmetric allylic alkylation
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

SPECIAL TOPIC
Asymmetric Ring-Opening Reaction of Oxabenzonorbornadienes
2109
Table 10 Substituted Oxabenzonorbornadienes
R1
R1
OH
HO
R1
R33Al (0.5–1 equiv)
CuX (3 mol%)
R3
R3
R1
+
+
O
L1 (3 mol%)
MTBE, r.t.
R2
R2
R2
R2
7, R1 = H, R2 = Me
8, R1 = R2 = Me
15
13
14
Entry
Substrate
R3
Product
13a
Conv. (%)
100
Ratioa 13/14/15
86:8:6
Ratio anti/syn
>99:1
eeb (%)
1
2
3
4
5
6
7c
7c
7c
7c
8d
8e
Me
Et
61
85
57
11
97
94
13b
100
26:52:22
19:14:63
60:12:28
6:3:84
>99:1
n-Pr
Ph
Et
13c
100
>99:1
13d
100
>99:1
13e
75
>99:1
Et
13e
77
44:17:39
>99:1
a Determined by 1H NMR.
b Determined by chiral SFC.
c Reaction realized with CuTC.
d Reaction realized with CuCl.
e Reaction realized with Cu(OAc)·H2O.
Copper-Catalyzed Allylic Alkylation with Aluminum Reagents;
General Procedure B
All reactions were carried out under an argon atmosphere with
oven-dried glassware. Solvents were dried by filtration over alumi-
na previously activated (350 °C, 12 h, N2) before use. All solvents
were degassed by N2 bubbling before use to all experiments.
Grignard reagents (in Et2O) and trialkylaluminum reagents (in hex-
ane) were used without any further purification. CuTC (Frontier
Scientific) was used without further purification. The reactions
were followed by GC-MS Hewlett Packard (EI mode) HP6890-
5973 or by TLC (visualization by UV and anisaldehyde, KMnO4 or
PMA staining). Flash chromatography was performed using silica
gel 32–63 mm, 60 Å. 1H (300, 400 or 500 MHz) and 13C (75, 100 or
125 MHz) NMR spectra were recorded in CDCl3 or C6D6 on Bruker
AMX-300, -400 or -500 spectrometers. Chemical shifts are relative
to residual deuterated solvent. Enantiomeric excess values were de-
termined by chiral GC measurement on either an HP6890 (H2 as
vector gas) or an HP6850 (H2 or He as vector gas) with the stated
column. Temperature programs are described as follows: initial
temperature (°C) – initial time (min) – temperature gradient (°C/
min) – final temperature (°C), gas flow (cm/s). In some cases, enan-
tiomeric excess values were determined by chiral SFC measure-
ment on a Berger SFC with the stated column. Gradient programs
are described as follows: initial methanol concentration (%) – initial
time (min) – percent gradient of methanol (%/min) – final methanol
concentration (%).
In a dried Schlenck tube under an N2 atmosphere were placed CuTC
(2.9 mg, 0.015 mmol, 3 mol%) and L1 (6.8 mg, 0.015 mmol, 3
mol%). MTBE was added (3 mL) and mixture was stirred at r.t. for
20 min. The oxabicyclic alkene (0.5 mmol) was added. R3Al in hex-
anes (0.6 mmol) was added dropwise over a period of 3 min. The
mixture was stirred for 20 h and then quenched with H2O and 1 M
HCl. The organic layer was separated and the aqueous phase was
extracted with CH2Cl2. The combined organic phases were dried
(MgSO4), filtered, and concentrated. The residue was purified by
flash chromatography.
(1S,2R)-trans-2-Ethyl-1,2-dihydronaphthalen-1-ol (2a)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 90% yield;
96% de; 82% ee (–20 °C); procedure B: 79% yield; 98% de; 92% ee.
Chiral separation: GC (Hydrodex B-6-TBDM, 100-0-1-170 47 cm/
s): tR1 = 44.0 min, tR2 = 45.3 min.
[a]D20 –264.2 (c 1.01, CHCl3, 82% ee).
1H NMR (400 MHz, CDCl3): d = 7.32 (dd, J = 6.6, 1.1 Hz, 1 H),
7.25–7.15 (m, 2 H), 7.06 (dd, J = 6.6, 1.3 Hz, 1 H), 6.45 (d, J = 9.5
Hz, 1 H), 5.96 (dd, J = 9.5, 4.8 Hz, 1 H), 4.49 (d, J = 4.8 Hz, 1 H),
2.48–2.39 (m, 1 H), 1.75 (1 H, OH), 1.49–1.22 (m, 2 H), 0.91 (t,
J = 7.3 Hz, 3 H).
Copper-Catalyzed Allylic Alkylation with Grignard Reagents;
General Procedure A
13C NMR (100 MHz, CDCl3): d = 135.8, 132.3, 130.7, 128.4, 127.7,
In a dried Schlenck tube under an N2 atmosphere were placed CuTC
(2.9 mg, 0.015 mmol, 3 mol%) and L1 (6.8 mg, 0.015 mmol, 3
mol%). CH2Cl2 was added (3 mL) and the mixture was stirred at r.t.
for 20 min. The oxabicyclic alkene (0.5 mmol) was added and the
system cooled down to the desired temperature if required. RMgBr
in Et2O (0.6 mmol) was added dropwise over a period of 3 min. The
mixture was stirred for 20 h and then quenched with MeOH and 1
M HCl before being allowed to reach r.t. The organic layer was sep-
arated and the aqueous phase was extracted with CH2Cl2. The com-
bined organic phases were dried (MgSO4), filtered, and
concentrated. The residue was purified by flash chromatography.
127.6, 126.4, 126.1, 72.1, 43.9, 24.5, 11.5.
(1S,2R)-trans-2-Methyl-1,2-dihydronaphthalen-1-ol (2b)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 84% yield;
80% de; 72% ee (r.t.); procedure B: 93% yield; 98% de; 91% ee.
Chiral separation: GC (Chirasil DEX-CB, 100-30-1-170 30 cm/s):
t
R1 = 64.4 min, tR2 = 68.2 min.
[a]D20 –242.6 (c 0.69, CHCl3, 91% ee).
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

2110
R. Millet et al.
SPECIAL TOPIC
1H NMR (400 MHz, CDCl3): d = 7.38 (d, J = 6.2 Hz, 1 H), 7.29–
7.19 (m, 2 H), 7.09 (d, J = 6.2 Hz, 1 H), 6.43 (d, J = 9.5 Hz, 1 H),
5.91 (dd, J = 9.5, 4.4 Hz, 1 H), 4.44 (d, J = 5.8 Hz, 1 H), 2.68–2.56
(m, 1 H), 1.74 (d, J = 5.8 Hz, 1 H), 1.05 (d, J = 6.9 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 135.7, 132.3, 128.3, 127.6, 127.2,
126.4, 125.8, 125.7, 74.1, 37.4, 16.9.
[a]D20 –205.9 (c 0.67, CHCl3, 64% ee).
1H NMR (400 MHz, CDCl3): d = 7.35 (m, 1 H), 7.26 (m, 2 H), 7.11
(m, 1 H), 6.50 (d, J = 9.9 Hz, 1 H), 6.03 (dd, J = 9.7, 4.8 Hz, 1 H),
5.09 (m, 1 H), 4.53 (d, J = 4.3 Hz, 1 H), 2.59 (m, 1 H), 2.07 (m, 2
H), 1.67 (s, 4 H, CH3, OH), 1.60 (s, 3 H), 1.43 (m, 1 H), 1.32 (m, 1
H).
13C NMR (100 MHz, CDCl3): d = 135.7, 132.5, 132.2, 131.0, 128.7,
128.1, 127.8, 127.8, 126.7, 126.1, 124.2, 72.6, 42.1, 31.8, 25.8,
25.6, 17.8.
(1S,2R)-trans-2-Butyl-1,2-dihydronaphthalen-1-ol (2c)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 86% yield;
96% de; 76% ee (–20 °C); procedure B: 95% yield; 98% de; 87% ee.
MS (EI): m/z = 228 (56), 157 (100), 144 (30), 128 (14), 115 (22), 91
(14), 55 (13).
Chiral separation: SFC (OD-H, 10%-2-1-15%): tR1 = 3.9 min, tR2
=
4.2 min.
(1S,2R)-trans-2-Isopropyl-1,2-dihydronaphthalen-1-ol (2g)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 65% con-
version; 90% de; 54% ee (–40 °C).
[a]D20 –286.3 (c 0.73, CHCl3, 87% ee).
1H NMR (400 MHz, CDCl3): d = 7.30 (d, J = 6.6 Hz, 1 H), 7.26–
7.14 (m, 2 H), 7.05 (d, J = 6.6 Hz, 1 H), 6.44 (d, J = 9.5 Hz, 1 H),
5.98 (dd, J = 9.5, 4.8 Hz, 1 H), 4.46 (d, J = 4.8 Hz, 1 H), 2.55–2.46
(m, 1 H), 1.76 (1 H, OH), 1.42–1.18 (m, 6 H), 0.83 (t, J = 7.0 Hz, 3
H).
13C NMR (100 MHz, CDCl3): d = 135.6, 132.3, 131.01, 128.4,
127.8, 127.5, 126.4, 125.8, 72.3, 42.4, 31.3, 29.2, 22.8, 13.9.
Chiral separation: SFC (IC-H, 2%-2-1-15%): tR1 = 7.3 min, tR2 = 8.0
min.
1H NMR (300 MHz, CDCl3): d = 7.34 (d, J = 6.9 Hz, 1 H), 7.31–
7.18 (m, 2 H), 7.11 (d, J = 7.7 Hz, 1 H), 6.55 (d, J = 9.5 Hz, 1 H),
5.97 (dd, J = 9.5, 4.8 Hz, 1 H), 4.65 (dd, J = 6.6, 4.4 Hz, 1 H), 2.46–
2.38 (m, 1 H), 1.74 (dq, J = 13.5, 6.6 Hz, 1 H), 1.63 (d, J = 6.6 Hz,
1 H, OH), 0.88 (d, J = 6.6 Hz, 3 H), 0.80 (d, J = 6.9 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 136.4, 128.8, 128.4, 127.7, 127.5,
126.8, 126.7, 126.5, 70.8, 41.9, 29.8, 20.7, 19.3.
(1S,2R)-trans-2-Isobutyl-1,2-dihydronaphthalen-1-ol (2d)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 73% con-
version, 90% de; 62% ee (–20 °C); procedure B: 71% yield; 98%
de; 94% ee.
(1S,2R)-trans-2-Phenyl-1,2-dihydronaphthalen-1-ol (2h)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 89% yield;
96% de; 0% ee (–20 °C); procedure B: 95% yield of a mixture of
phenyl and alkyl transfer; 98% de; 84% ee.
Chiral separation: GC (Chirasil DEX-CB, 120-0-1-170 30 cm/s):
tR1 = 37.1 min, tR2 = 47.9 min.
[a]D20 –345.9 (c 0.66, CHCl3, 94% ee).
1H NMR (300 MHz, CDCl3): d = 7.36–7.19 (m, 3 H), 7.12 (dd,
J = 7.0, 2.1 Hz, 1 H), 6.51 (d, J = 9.7 Hz, 1 H), 6.03 (dd, J = 9.7, 4.8
Hz, 1 H), 4.47 (t, J = 4.5 Hz, 1 H), 2.74–2.62 (m, 1 H), 1.82–1.63
(m, 2 H), 1.21–1.13 (m, 2 H), 0.93 (d, J = 4.8 Hz, 3 H), 0.90 (d,
J = 4.8 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 135.5, 132.3, 131.0, 128.6, 128.2,
127.6, 126.5, 125.7, 72.7, 40.7, 40.2, 25.5, 23.3, 22.1.
Chiral separation: SFC (OJ-H, 2%-2-1-15%): tR1 = 10.9 min, tR2
=
13.6 min.
1H NMR (300 MHz, CDCl3): d = 7.42 (dd, J = 7.3, 2.4 Hz, 1 H),
7.34–7.24 (m, 7 H), 7.18 (dd, J = 7.3, 2.4 Hz, 1 H), 6.67 (dd, J = 9.7,
2.0 Hz, 1 H), 6.04 (dd, J = 9.7, 3.6 Hz, 1 H), 4.84 (dd, J = 7.7, 5.3
Hz, 1 H), 3.83–3.79 (m, 1 H), 1.96 (d, J = 5.3 Hz, 1 H).
13C NMR (100 MHz, CDCl3): d = 140.8, 135.6, 132.6, 129.8, 128.8
(1S,2R)-trans-2-Phenethyl-1,2-dihydronaphthalen-1-ol (2e)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 64% yield;
96% de; 57% ee (–20 °C).
(2 C), 128.4, 128.2, 128.1, 127.6, 127.2, 126.4, 74.3, 50.1.
(1S,2R)-trans-2-Propyl-1,2-dihydronaphthalen-1-ol (2i)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure B: 50% yield;
98% de; 87% ee.
Chiral separation: SFC (OJ-H, 10%-2-1-15%): tR1 = 8.9 min, tR2
=
10.0 min.
Chiral separation: SFC (OD-H, 10%-2-1-15%): tR1 = 3.8 min, tR2
4.1 min.
[a]D20 –323.7 (c 0.49, CHCl3, 87% ee).
1H NMR (400 MHz, CDCl3): d = 7.35 (dd, J = 6.1, 0.7 Hz, 1 H),
7.26 (m, 3 H), 7.11 (dd, J = 7, 1.3 Hz, 1 H), 6.50 (d, J = 10.6 Hz, 1
H), 6.02 (dd, J = 10.6, 4.8 Hz, 1 H), 4.52 (d, J = 4.3 Hz, 1 H), 2.50
(m, 1 H), 1.57 (br, 1 H), 1.43–1.34 (m, 3 H), 1.27 (m, 1 H), 0.90 (t,
J = 7.1 Hz, 1 H).
13C NMR (100 MHz, CDCl3): d = 135.7, 132.5, 131.1, 128.7, 128.1,
127.8, 126.6, 126.0, 72.6, 42.4, 34.0, 20.4, 14.4.
=
[a]D20 –198.5 (c 0.77, CHCl3, 64% ee).
1H NMR (400 MHz, CDCl3): d = 7.35 (m, 1 H), 7.26 (m, 4 H), 7.16
(m, 4 H), 6.55 (d, J = 9.6 Hz, 1 H), 6.08 (dd, J = 9.6, 4.8 Hz, 1 H),
4.56 (d, J = 4.3 Hz, 1 H), 2.78–2.59 (m, 3 H), 1.81–1.57 (m, 3 H).
13C NMR (100 MHz, CDCl3): d = 142.1, 135.6, 132.4, 130.5, 128.7,
128.5, 128.4, 128.0, 127.9, 126.7, 126.5, 126.0, 72.5, 42.0, 33.6,
33.4.
MS (EI): m/z = 250 (21), 232 (39), 159 (16), 141 (100), 131 (30),
115 (28), 105 (14), 91 (61), 65 (13).
(1S,2R)-trans-2-(4-Methylpent-3-enyl)-1,2-dihydronaphthalen-
1-ol (2f)
MS (EI): m/z = 188 (28), 159 (35), 145 (100), 117 (27), 91 (17), 51
(6).
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 95% yield;
94% de; 64% ee (–20 °C).
(1S,2R)-trans-2-Ethyl-6,7-difluoro-1,2-dihydronaphthalen-1-ol
(9a)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure A: 85% yield;
92% de; 64% ee.
Chiral separation: SFC (OD-H, 5%-2-1-15%): tR1 = 6.6 min, tR2
7.1 min.
=
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

SPECIAL TOPIC
Asymmetric Ring-Opening Reaction of Oxabenzonorbornadienes
2111
Chiral separation: GC (Hydrodex B-6-TBDM, 90-0-1-170 47 cm/
1 H), 3.45 (s, 3 H), 3.41 (s, 3 H), 2.51 (m, 1 H), 0.97 (d, J = 5.2 Hz,
s): tR1 = 63.4 min, tR2 = 64.1 min.
3 H).
[a]D20 –132.2 (c 0.50, CHCl3, 64% ee).
13C NMR (100 MHz, CDCl3): d = 149.73, 149.71, 130.7, 129.9,
127.9, 126, 111.8, 111.1, 74.5, 55.8, 55.7, 38.2, 17.5.
1H NMR (400 MHz, CDCl3): d = 7.21 (dd, J = 10.4, 8.1 Hz, 1 H),
6.90 (dd, J = 10.4, 7.8 Hz, 1 H), 6.39 (d, J = 9.6 Hz, 1 H), 5.99 (dd,
J = 9.6, 4.5 Hz, 1 H), 4.49 (d, J = 5.8 Hz, 1 H), 2.45 (m, 1 H), 1.82
(br, 1 H), 1.54–1.47 (m, 1 H), 1.40–1.33 (m, 1 H), 0.97 (t, J = 7.3
Hz, 3 H).
MS (EI): m/z = 220 (95), 202 (100), 178 (13), 159 (22), 115 (22), 91
(11), 77 (11), 51 (11).
HRMS (EI): m/z calcd for C13H16O3: 220.1099; found: 220.1096.
13C NMR (100 MHz, CDCl3): d = 150 (m), 148 (m), 133 (m), 131.4,
129.5 (m), 124.7, 116.6 (d, J = 19 Hz), 115.1 (d, J = 19 Hz), 43.8,
24.4, 11.4.
19F NMR (270 MHz, CDCl3): d = – 138.85 (dd, J = 21.7, 3.1 Hz),
–139.65 (dd, J = 21.7, 3.1 Hz).
(1S,2R)-trans-2-Ethyl-5,8-dimethoxy-1,2-dihydronaphthalen-1-
ol (11a)
The crude could not be purified by flash chromatography (silica gel)
due to decomposition; procedure B: 100% conversion; 98% de;
88% ee.
Chiral separation: SFC (AD-H, 5%-2-1-15%): tR1 = 8.1 min, tR2
=
(1S,2R)-trans-6,7-Difluoro-2-methyl-1,2-dihydronaphthalen-1-
ol (9b)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure B: 85% yield;
88% de; 91% ee.
8.7 min.
1H NMR (400 MHz, CDCl3): d = 6.88 (dd, J = 9.9, 0.5 Hz, 1.H),
6.78 (d, J = 9.1 Hz, 1 H), 6.78 (d, J = 8.8 Hz, 1 H), 6.08 (ddd,
J = 9.9, 5.6, 1.3 Hz, 1 H), 4.99 (br, 1 H), 3.83 (s, 3 H), 3.80 (s, 3 H),
2.53 (m, 1 H), 1.58 (br, 1 H), 1.34–1.25 (m, 2 H), 0.94 (t, J = 7.6 Hz,
3 H).
[a]D20 –169.5 (c 0.56, CHCl3, 91% ee).
Chiral separation: GC (Hydrodex B-6-TBDM, 90-0-1-170 47 cm/
s): tR1 = 63.4 min, tR2 = 64.1 min.
(1S,2R)-trans-5,8-Dimethoxy-2-methyl-1,2-dihydronaphthalen-
1-ol (11b)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15) as a white solid; proce-
dure B: 88% yield; 86% de; 87% ee.
1H NMR (400 MHz, CDCl3): d = 7.26 (m, 2 H), 6.90 (dd, J = 10.6,
7.6 Hz, 1 H), 6.34 (dd, J = 9.6, 1.5 Hz, 1 H), 5.92 (dd, J = 9.6, 4 Hz,
1 H), 4.41 (d, J = 7.3 Hz, 1 H), 2.59 (m, 1 H), 1.64 (br, 1 H), 1.09
(d, J = 7.3 Hz, 1 H).
13C NMR (100 MHz, CDCl3): d = 151.5 (dd, J = 41.2, 12.2 Hz),
148.1 (dd, J = 41.8, 12.2 Hz), 133.2, 133.1 (m), 129.5 (m), 124,
116.2 (d, J = 18 Hz), 115.0 (d, J = 18 Hz), 73.6, 37.4, 17.1.
Chiral separation: SFC (AS-H, 5%-2-1-15%): tR1 = 4.1 min, tR2
=
5.6 min.
[a]D20 –223.3 (c 0.76, CHCl3, 87% ee).
1H NMR (400 MHz, CDCl3): d = 7.14 (dd, J = 0.8 Hz, 1 H), 6.48
(d, J = 9.1 Hz, 1 H), 6.44 (d, J = 8.8 Hz, 1 H), 5.89 (ddd, J = 9.8,
5.3, 1 Hz, 1 H), 5.14 (d, J = 1.8 Hz, 1 H), 3.36 (s, 3 H), 3.34 (s, 3
H), 2.70 (m, 1 H), 2.14 (br, 1 H), 0.86 (s, J = 7.3 Hz, 3 H).
19F NMR (270 MHz, CDCl3) : d = –138.85 (dd, J = 21.7, 3.1 Hz),
–139.65 (dd, J = 21.7, 3.1 Hz).
(1S,2R)-trans-2-Ethyl-6,7-dimethoxy-1,2-dihydronaphthalen-1-
ol (10a)
13C NMR (100 MHz, CDCl3): d = 152.1, 150.0, 132.2, 124.9, 122.8,
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15) as a yellow oil; procedure
A: 50% yield; >90% de; 31% ee (–40 °C); procedure B: 95% yield;
98% de; 89% ee.
119.1, 110.9, 110.5, 67.3, 55.63, 55.59, 36.9, 17.3.
(1S,2R)-trans-2,4-Dimethyl-1,2-dihydronaphthalen-1-ol (13a)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure B: 20% yield;
99% de; 61% ee.
Chiral separation: SFC (AD-H, 5%-2-1-15%): tR1 = 8.4 min, tR2
=
10.2 min.
Chiral separation: GC (CP-Chirasil-DEX CB, 100-30-1-155-0-20-
170-2 30 cm/s): tR1 = 74.03 min, tR2 = 75.10 min.
1H NMR (300 MHz, CDCl3): d = 7.45–7.43 (d, J = 7.1 Hz, 1 H),
7.31–7.27 (m, 3 H), 5.7 (d, J = 2.8 Hz, 1 H), 4.44–4.41 (t, J = 13.9,
6.8 Hz, 1 H), 2.61–2.53 (m, 1 H), 2.08 (s, 3 H), 1.75–1.73 (d, J = 7.4
Hz, 1 H, OH), 1.07–1.05 (d, J = 7.1 Hz, 3 H).
[a]D20 –175.7 (c 0.69, CHCl3, 89% ee).
1H NMR (400 MHz, CDCl3): d = 6.90 (s, 1 H), 6.48 (s, 1 H), 6.34
(d, J = 9.6 Hz, 1 H), 5.79 (dd, J = 9.6, 4.29 Hz, 1 H), 4.46 (d, J = 6.1
Hz, 1 H), 3.44 (s, 3 H), 3.41 (s, 3 H), 2.39 (m, 1 H), 1.47–1.25 (m,
3 H), 0.88 (t, J = 7.3 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 149.8, 149.7, 129.9, 128.8, 128.6,
126.3, 126, 112.3, 111.2, 72.3, 55.8, 44.7, 25, 11.6.
13C NMR (100 MHz, CDCl3): d = 136.90, 134.35, 130.80, 129.42,
128.49, 127.7, 127.1, 123.59, 74.87, 37.65, 19.42, 17.6.
(1S,2R)-trans-6,7-Dimethoxy-2-methyl-1,2-dihydronaphthalen-
1-ol (10b)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15) as a white solid; proce-
dure A: 40% yield; 90% de; 20% ee (–20 °C); procedure B: 69%
yield; 98% de; 88% ee.
(1S,2R)-trans-2-Ethyl-4-methyl-1,2-dihydronaphthalen-1-ol
(13b)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure B: 22% yield;
99% de; 85% ee.
Chiral separation: SFC (Chiralpak AD: 5%-2-1-15%): tR1 = 5.43
min, tR2 = 6.25 min.
1H NMR (300 MHz, CDCl3): d = 7.57–7.44 (m, 4 H), 5.72 (m, 1 H),
4.52 (d, J = 4.8 Hz, 1 H), 2.48 (m, 1 H), 2.1 (s, 3 H), 1.5–1.4 (m, 2
H), 0.95 (s, 3 H).
Chiral separation: SFC (AD-H, 5%-2-1-15%): tR1 = 9.3 min, tR2
10.3 min.
[a]D20 –185.5 (c 0.74, CHCl3, 88% ee).
1H NMR (400 MHz, CDCl3): d = 6.96 (s, 1 H), 6.48 (s, 1 H), 6.30
(dd, J = 9.5, 1.5 Hz, 1 H), 5.68 (dd, J = 9.5, 4.1 Hz, 1 H), 4.34 (m,
=
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York

2112
R. Millet et al.
SPECIAL TOPIC
(1S,2R)-trans-4-Methyl-2-propyl-1,2-dihydronaphthalen-1-ol
(13c)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 85:15); procedure B: <10%
yield; 99% de; 57% ee.
(5) (a) Lautens, M.; Renaud, J.-L.; Hiebert, S. J. Am. Chem. Soc.
2000, 122, 1804. (b) Lautens, M.; Hiebert, S.; Renaud, J.-L.
Org. Lett. 2000, 2, 1971. (c) Lautens, M.; Hiebert, S.;
Renaud, J.-L. J. Am. Chem. Soc. 2001, 123, 6834.
(d) Priego, J.; García Mancheño, O.; Cabrera, S.;
Gómez Arrayás, R.; Llamas, T.; Carretero, J. C. Chem.
Commun. 2002, 2512. (e) Lautens, M.; Hiebert, S. J. Am.
Chem. Soc. 2004, 126, 1437. (f) Cabrera, S.;
Chiral separation: SFC (Chiralcel OD-H: 5%-2-1-15%): tR1 = 5.93
min, tR2 = 6.33 min.
1H NMR (300 MHz, CDCl3): d = 7.40–7.28 (m, 4 H), 5.82–5.81 (d,
J = 4.8 Hz, 1 H), 4.50–4.47 (t, J = 12.1, 6.3 Hz, 1 H), 2.51–2.54 (m,
1 H), 2.09 (s, 3 H), 1.75–1.73 (d, J = 7.3 Hz, 1 H, OH), 1.44–1.25
(dm, 4 H), 0.91–0.88 (t, J = 14.4, 7.1 Hz, 3 H).
13C NMR (100 MHz, CDCl3): d = 136.66, 134.25, 130.82, 128.74,
127.95, 127.90, 127.79, 123.69, 73.17, 42.39, 34.25, 20.60, 19.56,
14.60.
Gómez Arrayás, R.; Carretero, J. C. Angew. Chem. Int. Ed.
2004, 43, 3944. (g) Dotta, P.; Kuwar, P. G. A.; Pregosin, P.
S.; Albinati, A.; Rizzato, S. Organometallics 2004, 23,
2295. (h) Li, M.; Yan, X.-X.; Hong, W.; Zhu, X.-Z.; Cao,
B.-X.; Sun, J.; Hou, X.-L. Org. Lett. 2004, 6, 2833.
(i) Imamoto, T.; Sugita, K.; Yoshida, K. J. Am. Chem. Soc.
2005, 127, 11934. (j) Cabrera, S.; Gómez Arrayás, R.;
Alonso, I.; Carretero, J. C. J. Am. Chem. Soc. 2005, 127,
17938.
(1S,2R)-trans-4-Methyl-2-phenyl-1,2-dihydronaphthalen-1-ol
(13d)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 90:10); procedure B: 16% yield;
99% de; 11% ee.
(6) Millward, D. B.; Sammis, G.; Waymouth, R. M. J. Org.
Chem. 2000, 65, 3902.
(7) (a) Murakami, M.; Igawa, H. Chem. Commun. 2002, 390.
(b) Lautens, M.; Dockendorff, C.; Fagnou, K.; Malicki, A.
Org. Lett. 2002, 4, 1311. (c) Webster, R.; Bing, Ch.;
Lautens, M. J. Am. Chem. Soc. 2009, 131, 444.
(8) Zhang, T.-K.; Mo, D.-L.; Dai, L.-X.; Hou, X.-L. Org. Lett.
2008, 10, 3689.
(9) (a) Lautens, M.; Fagnou, K. J. Am. Chem. Soc. 2001, 123,
7170. (b) Lautens, M.; Schmid, G. A.; Chau, A. J. Org.
Chem. 2002, 67, 8043. (c) Lautens, M.; Fagnou, K.; Yang,
D. J. Am. Chem. Soc. 2003, 125, 14884.
(10) Bertozzi, F.; Pineschi, M.; Macchia, F.; Arnold, L. A.;
Minnaard, A. J.; Feringa, B. L. Org. Lett. 2002, 4, 2703.
(11) (a) Gomez Arrayas, R.; Cabrera, S.; Carretero, J. C. Org.
Lett. 2003, 5, 1333. (b) Gomez Arrayas, R.; Cabrera, S.;
Carretero, J. C. Synthesis 2006, 1205.
(12) (a) Zhang, W.; Wang, L.-X.; Shi, W.-J.; Zhou, Q.-L. J. Org.
Chem. 2005, 70, 3734. (b) Xie, J.-H.; Zhou, Q.-L. Acc.
Chem. Res. 2008, 41, 581. (c) Zhang, W.; Zhu, S.-F.; Qiao,
X.-Ch.; Zhou, Q.-L. Chem. Asian J. 2008, 3, 2105.
(13) Palais, L.; Mikhel, I. S.; Bournaud, C.; Micouin, L.; Falciola,
C. A.; Vuagnoux-d’Augustin, M.; Rosset, S.; Bernardinelli,
G.; Alexakis, A. Angew. Chem. Int. Ed. 2007, 46, 7462.
(14) Millet, R.; Alexakis, A. Synlett 2008, 1797.
(15) (a) Alexakis, A.; Malan, Ch.; Lea, L.; Tissot-Croset, K.;
Polet, D.; Falciola, C. Chimia 2006, 60, 124. (b) Falciola,
C.; Alexakis, A. Eur. J. Org. Chem. 2008, 3765.
(16) Clavier, H.; Coutable, L.; Toupet, L.; Guillemin, J.-C.;
Mauduit, M. J. Organomet. Chem. 2005, 690, 5237.
(17) Alexakis, A.; Benhaim, C.; Rosset, S.; Humam, M. J. Am.
Chem. Soc. 2002, 124, 5262.
Chiral separation: SFC (Chiralcel OD-H: 5%-2-1-15%): tR1 = 9.69
min, tR2 = 10.45 min.
1H NMR (300 MHz, CDCl3): d = 7.37–7.23 (m, 9 H), 5.95 (s, 1 H),
4.92–4.88 (t, J = 13.9, 6.3 Hz, 1 H), 3.84 (s, 1 H), 2.19 (s, 3 H),
1.49–1.47 (d, J = 8.4 Hz, 1 H, OH).
(1S,2R)-trans-2-Ethyl-1,4-dimethyl-1,2-dihydronaphthalen-1-
ol (13e)
The crude product was purified by flash chromatography (silica gel,
pentane then cyclohexane–EtOAc, 90:10); procedure B: 35% yield;
99% de; 97% ee.
Chiral separation: SFC (Chiralpak AD: 5%-2-1-15%): tR1 = 4.57
min, tR2 = 5.65 min.
1H NMR (300 MHz, CDCl3): d = 7.61 (m, 1 H), 7.35–7.26 (m, 3 H),
5.75 (s, 1 H), 2.30 (m, 1 H), 2.10 (s, 3 H), 1.9–1.8 (m, 2 H), 1.32 (s,
3 H), 1.0 (t, J = 7.2 Hz, 3 H).
References
(1) (a) Willis, M. C. J. Chem. Soc., Perkin Trans. 1 1999, 1765.
(b) Rovis, T. In New Frontiers in Asymmetric Catalysis;
Mikami, K.; Lautens, M., Eds.; Wiley: New Jersey, 2007,
275–311.
(2) (a) Lautens, M. Synlett 1993, 177. (b) Chiu, P.; Lautens, M.
Top. Curr. Chem. 1997, 190, 1. (c) Lautens, M.; Fagnou,
K.; Hiebert, S. Acc. Chem. Res. 2003, 36, 48.
(18) Marino, J. P.; Floyd, D. M. Tetrahedron Lett. 1979, 20, 675.
(19) Bournaud, C.; Falciola, C.; Lecourt, T.; Rosset, S.; Alexakis,
A.; Micouin, L. Org. Lett. 2006, 8, 3581.
(20) Zweifel, G.; Miller, J. A. Org. React. 1984, 32, 375.
(21) Hawner, C.; Li, K.; Cirriez, V.; Alexakis, A. Angew. Chem.
Int. Ed. 2008, 47, 8211.
(3) (a) Caple, R.; Chen, G. M. S.; Nelson, J. D. J. Org. Chem.
1971, 36, 2874. (b) Arjona, O.; Fernandez de la Pradilla, R.;
Garcia, E.; Martin-Domenech, A.; Plumet, J. Tetrahedron
Lett. 1989, 30, 6437. (c) Lautens, M.; Gajda, C.; Chiu, P.
J. Chem. Soc., Chem. Commun. 1993, 1193.
(4) Lautens, M.; Di Felice, C.; Huboux, A. Tetrahedron Lett.
1989, 30, 6817.
Synthesis 2009, No. 12, 2101–2112 © Thieme Stuttgart · New York