Desymmetrization of 1,4-cyclohexadienyltriisopropoxysilane using copper catalysis

Article abstract of DOI:10.1021/ol070689d

The first catalytic desymmetrization in the field of allylsilane chemistry is presented. Desymmetrization of cyclohexadienyltriisopropoxysilane is achieved using coppper catalysis. High diastereo- and enantioselectivities are obtained, and the product die

Full text of DOI:10.1021/ol070689d

ORGANIC
LETTERS
2007
Vol. 9, No. 11
2175-2178
Desymmetrization of
1,4-Cyclohexadienyltriisopropoxysilane
Using Copper Catalysis
Rui Umeda and Armido Studer*
Organisch-Chemisches Institut, Westfa¨lische Wilhelms-UniVersita¨t, Corrensstrasse 40,
48149 Mu¨nster, Germany
studer@uni-muenster.de
Received March 21, 2007
ABSTRACT
The first catalytic desymmetrization in the field of allylsilane chemistry is presented. Desymmetrization of cyclohexadienyltriisopropoxysilane
is achieved using coppper catalysis. High diastereo- and enantioselectivities are obtained, and the product dienes are highly valuable building
blocks for natural product synthesis.
Various allylmetal compounds have been successfully used
for the asymmetric allylation of aldehydes.1 Allylsilanes have
often been applied as nucleophiles in these reactions. They
are stable compounds, and some of them are commercially
available. Importantly, allylsilanes are nontoxic and generally
stable toward moisture. However, most of the allylsilanes
are not reactive enough to undergo spontaneous addition to
an aldehyde, and Lewis acid activation of the aldehyde is
necessary.2 Activation of allylsilanes can also be achieved
by Lewis bases.3 A third option is the transmetalation of an
allylsilane to generate a reactive allylmetal species that is
able to react with an aldehyde. In fact, Cu(I)-4 and Ag(I)-
catalyzed5 enantioselective allylations of various aldehydes
using readily available allyltrialkoxysilanes have been pub-
lished. Achiral Cd(II) complexes have also been reported to
catalyze the allylation of aldehydes using trialkoxyallylsi-
lanes.6 Moreover, hydrazones and activated imines were
successfully allylated by allylsilanes using chiral Zn7 and
Cu complexes.8
We have recently reported the desymmetrization9 of 1,4-
cyclohexadiene using the chiral Ti(IV) complex 1. The
synthetically highly valuable 1,3-cyclohexadienes 2 were
obtained with high diastereo- and enantioselectivities (Scheme
1).10 In these reactions the differentiation of the two
(5) (a) Yanagisawa, A.; Kageyama, H.; Nakatsuka, Y.; Asakawa, K.;
Matsumoto, Y.; Yamamoto, H. Angew. Chem., Int. Ed. 1999, 38, 3701. (b)
Wadamoto, M.; Ozasa, N.; Yanagisawa, A.; Yamamoto, H. J. Org. Chem.
2003, 68, 5593. (c) Wadamoto, M.; Yamamoto, H. J. Am. Chem. Soc. 2005,
127, 14556.
(6) (a) Aoyama, N.; Hamada, T.; Manabe, K.; Kobayashi, S. Chem.
Commun. 2003, 676. (b) Aoyama, N.; Hamada, T.; Manabe, K.; Kobayashi,
S. J. Org. Chem. 2003, 68, 7329.
(7) Hamada, T.; Manabe, K.; Kobayashi, S. Angew. Chem., Int. Ed. 2003,
42, 3927.
(8) Kiyohara, H.; Nakamura, Y.; Matsubara, R.; Kobayashi, S. Angew.
Chem., Int. Ed. 2006, 45, 1615.
(9) (a) Studer, A.; Schleth, F. Synlett 2005, 3033. (b) Hoffman, R. W.
Synthesis 2004, 2075. (c) Rahman Abd, N.; Landais, Y. Curr. Org. Chem.
2002, 6, 1369. (c) Willis, M. C. J. Chem. Soc., Perkin Trans. 1 1999, 1765.
(10) (a) Schleth, F.; Studer, A. Angew. Chem., Int. Ed. 2004, 43, 313.
(b) Schleth, F.; Vogler, T.; Harms, K.; Studer, A. Chem. Eur. J. 2004, 10,
4171.
(1) (a) In Houben-Weyl; Helmchen, G., Hoffmann, R. W., Mulzer, J.,
Schaumann, E., Eds.; Thieme: Stuttgart, 1995; Vol. E21 b, pp 1357-1602.
(b) Yamamoto, Y.; Asao, N. Chem. ReV. 1993, 93, 2207. (c) Denmark, S.
E.; Fu, J. Chem. ReV. 2003, 103, 2763.
(2) (a) Fleming, I.; Barbero, A.; Walter, D. Chem. ReV. 1997, 97, 2063.
(b) Thomas, E. J. Allylsilanes. In Houben-Weyl; Helmchen, G., Hoffmann,
R. W., Mulzer, J., Schaumann, E., Eds.; Thieme: Stuttgart, 1995; Vol. E21
b, p 1491.
(3) (a) Malkov, A. V.; Dufkova´, L.; Farrugia, L.; Kocˇovsky´, P. Angew.
Chem., Int. Ed. 2003, 42, 3674. (b) Denmark, S. E.; Fu, J. Chem. Commun.
2003, 167. (c) Berger, R.; Rabbat, P. M. A.; Leighton, J. L. J. Am. Chem.
Soc. 2003, 125, 9596.
(4) (a) Yamasaki, S.; Fujii, K.; Wada, R.; Kanai, M.; Shibasaki, M. J.
Am. Chem. Soc. 2002, 124, 6536. (b) Itoh, T.; Miyazaki, M.; Fukuoka, H.;
Nagata, K.; Ohsawa, A. Org. Lett. 2006, 8, 1295.
10.1021/ol070689d CCC: $37.00
© 2007 American Chemical Society
Published on Web 04/26/2007

Scheme 1. Stoichiometric and Catalytic Desymmetrization of
Table 1. Screening of Various Metal Salts as Catalysts Using
Benzaldehyde as Electrophile (R ) Ph)
1,4-Cyclohexadienes
entry
catalyst
time (h) ratio (syn-2:anti-2:4) yield (%)a
1b
2
3
4
5
6
7
8
9
20
2
2
43:07:50
46:13:41
42:15:43
41:25:34
76:08:16
54:24:22
41:32:27
43:04:53
30:13:57
44:11:45
32:17:51
12:09:79
29
85
91
75
88
47
47
26
54
38
30
38
AgF
AgOTf
Cu(OTf)2
CuOTfc
CuBr
20
20
20
20
20
20
20
20
20
CuI
Zn(OTf)2
Yb(OTf)3
InBr3
enantiotopic double bonds (desymmetrization of the 1,4-
cyclohexadiene) was achieved and the diastereoselectivity
was perfectly controlled. A drawback of the developed
method is the fact that stoichiometric amounts of the chiral
complex 1 have to be used. Herein we present first results
on the catalytic desymmetrization of 1,4-cyclohexadienes.
As a suitable substrate to run the desymmetrization
catalytically, we identified the silylated cyclohexadiene 3,
which is readily prepared on a multigram scale by Birch
reduction of triisopropoxysilylbenzene.11 Benzaldehyde was
used as electrophile to screen various catalysts. The reactions
were performed in THF using 3 (2 equiv), MeOH (1 equiv),
a catalyst (10 mol %), PPh3 (40 mol %), and tetrabutylam-
monium fluoride (TBAF, 1 equiv) at room temperature. The
products syn-2, anti-2, and 4 were isolated by chromatog-
10
11b Ru(PPh3)3Cl2
12b Rh(PPh3)3Cl
a Combined yield. b The reaction was carried out without PPh3. c [CuOTf]2-
benzene complex was used.
these initial studies, AgOTf, Cu(I) salts, and Cu(OTf)2 were
identified as the best catalysts.Various chiral ligands 5-10
were then tested in the desymmetrization of 3 (Figure 1).
1
raphy on SiO2, and the isomer ratio was determined by H
NMR spectroscopy and GC analysis (Scheme 2, Table 1).
Scheme 2. Desymmetrization of the Cyclohexadiene 3
Importantly, the background reaction induced by TBAF
without any catalyst was slow (entry 1). Pleasingly, AgF in
combination with PPh3 was able to catalyze the “allylation”,
and the products were obtained in 85% combined yield as a
mixture of the three isomers (entry 2). The best result was
obtained with AgOTf (entry 3). A lower but still satisfactory
yield was achieved using Cu(OTf)2 (entry 4). However, as
compared to the Ag+-catalyzed processes, the reaction took
longer. An improvement of the yield and the selectivity
resulted for the CuOTf-catalyzed reaction (entry 5). Surpris-
ingly, worse results were obtained for the CuBr- and CuI-
mediated reactions (entries 6 and 7). The addition of
Zn(OTf)2 did not show any effect (entry 8). Moderate yields
were obtained for the Yb(OTf)3- and InBr3-catalyzed reac-
tions (entries 9 and 10). For all of these processes, Lewis
acid type catalysis cannot be excluded. Addition of Ru or
Rh catalysts revealed some effects on the isomer ratio;
however, low yields were obtained (entries 11 and 12). From
Figure 1. Chiral ligands 5-10 used in this study.
The reactions were run in THF at -25 °C for 20 h using 3
(2 equiv), 10 mol % of the Ag or Cu salt, and the chiral
ligand (10 mol %) in the presence of TBAF (1 equiv) and
MeOH (1 equiv). The enantiomeric ratio (er) of the syn-
isomer was determined by GC (see Supporting Informa-
tion).12
Excellent yields were obtained for all of the AgOTf-
catalyzed reactions (Table 2). Unfortunately, the undesired
regioisomer 4 was formed as the major product using
BINAP, 6, 8, and 9 (entries 1, 2, 4, and 5). The best result
in the AgOTf series was achieved using ligand 7, where 2
(11) Kammel, T.; Pachaly, B.; Weidner, R.; Studer, A.; Alvarez-Garcia,
L.; Herrera-Gonzalez, A. DE Patent 10326577 A, 2004.
(12) The assignment of the absolute configuration of the syn isomer is
based on literature data; see ref 10.
2176
Org. Lett., Vol. 9, No. 11, 2007

diminishing the selectivity (entry 24). The er was improved
upon switching to ligand 6 (entry 25). Reaction with CuOTf
under the same conditions gave the same result, supporting
the notion that a Cu(I) species is the active catalyst (entry
26).13 Importantly, the reaction is not moisture-sensitive and
can be performed using THF of lower quality (technical
grade) without affecting the yield and selectivity.
Table 2. Screening of Various Chiral Ligands L (R ) Ph; see
Scheme 2)
entry catalyst
L
ratio (syn-2:anti-2:4) er syn-2 yield (%)a
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
AgOTf
CuBr
CuI
CuOTfb
CuOTfb
CuOTfb
CuOTfb
CuOTfb
CuOTfb
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
Cu(OTf)2
5
6
7
8
9
10
5
5
5
6
7
8
9
10
5
6
7
8
9
31:09:60
29:08:63
74:05:21
22:22:56
36:18:46
50:05:45
43:10:47
42:09:49
79:13:08
80:11:09
88:09:03
88:04:08
68:17:15
83:11:06
88:10:02
88:11:01
83:11:06
76:09:15
46:14:40
63:08:29
89:09:02
90:08:02
89:09:02
89:09:02
90:08:02
86:11:03
71:29
77:23
77:23
49:51
40:60
41:59
57:43
54:46
86:14
92:08
84:16
74:26
46:54
22:78
86:14
94:06
83:17
68:32
47:53
28:72
86:14
85:15
86:14
86:14
94:06
94:06
87
95
99
99
99
99
38
59
26
36
36
21
31
36
32
38
54
43
36
32
38
44
60
83
83
83
Scheme 3. Suggested Mechanism (A and B Models for
R-DIFLUORPHOS as a Ligand)
Cu(OTf)2 10
21c Cu(OTf)2
22d Cu(OTf)2
23e Cu(OTf)2
24f
25f
26f
5
5
5
5
6
6
Cu(OTf)2
Cu(OTf)2
CuOTfb
The suggested mechanism is depicted in Scheme 3. A
similar mechanism was suggested by Shibasaki for the Cu-
mediated allylation of aldehydes using trimethoxyallyl
silane.4a TBAF-induced desilylation of 3 in the presence of
13, which we assume to be the resting state of the catalyst,
leads to the chiral Cu(I) complex 11. Reaction with RCHO
provides Cu-alcoholate 12, which is eventually hydrolyzed
with CF3CH2OH to give 2 and 13. We believe that the
geometry of the phenyl groups at the P-atom in the Cu(I)
complex resembles the geometry of the phenyl groups in
BINAP-Ru complexes.14 The selectivities can be explained
using the quadrant-shielding model of BINAP-metal com-
plexes introduced by Noyori.15 Indeed, this model has already
been used to explain selectivities in Cu(I)-BINAP-catalyzed
stereoselective reactions.16 Therefore, we assume that in order
to minimize the interactions of the aldehyde with the axial
phenyl group at the P-atom in the (R)-DIFLUORPHOS-Cu-
(I) complex the aldehyde approaches the metal as suggested
in model structure A. Hence, intramolecular “allylation”
using (R)-DIFLUORPHOS as ligand occurs from the re face
as observed in the experiment. The relative syn configuration
obtained in the experiment can be understood assuming a
six-membered chair transition state as suggested in model
B.
a Combined yield. b [CuOTf]2-benzene complex was used. c Three equiva-
lents of MeOH were added. d CF3CH2OH (1 equiv) instead of MeOH.
e CF3CH2OH (3 equiv) instead of MeOH. f CF3CH2OH (3 equiv) instead
of MeOH using 3 equiv of 3.
was formed with high syn:anti selectivity but moderate
enantioselectivity (entry 3). The CuBr- and CuI-mediated
reactions using BINAP afforded only low levels of enanti-
oselectivity (entries 7 and 8). As compared to the Ag-
catalyzed reactions, CuOTf catalysis gave lower yields but
better selectivities. BINAP and 7 provided similar results
(entries 9 and 11), and 6 turned out to be the best ligand
(entry 10): syn-2 was obtained with high enantioselectivity
(er ) 92:8) and good syn:anti selectivity, and regioisomer 4
was formed in only small amounts. Worse results were
achieved using 8, 9, or 10 (entries 12-14). Replacing CuOTf
with Cu(OTf)2 provided similar enantioselectivities (entries
16-20). Probably, the Cu(II)salt is reduced by the phosphine
ligands during the reaction.13
We next tried to improve the yield of the Cu(OTf)2-
catalyzed reaction. Increasing the amount of MeOH to 3
equiv provided a slightly higher yield without affecting the
enantioselectivity (entry 21). Replacing MeOH with 1 equiv
of CF3CH2OH afforded a higher yield (entry 22), which was
further increased to 60% upon using 3 equiv of CF3CH2OH
(entry 23) and to 83% if 3 equiv of 3 was used without
Finally, various aldehydes were tested in the desymme-
trization of 3 (Table 3). The regioisomer 4 was formed in
(14) Ohta, T.; Takaya, H.; Noyori, R. Inorg. Chem. 1988, 27, 566.
(15) Kitamura, M.; Tsukamoto, M.; Bessho, Y.; Yoshimura, M.; Kobs,
U.; Widhalm, M.; Noyori, R. J. Am. Chem. Soc. 2002, 124, 6649.
(16) (a) Yamamoto, Y.; Yamamoto, H. J. Am. Chem. Soc. 2004, 126,
4128. (b) Yamamoto, Y.; Yamamoto, H. Angew. Chem., Int. Ed. 2005, 44,
7082.
(13) (a) Kru¨ger, J.; Carreira, E. M. J. Am. Chem. Soc. 1998, 120, 837.
(b) Pagenkopf, B. L.; Kru¨ger, J.; Stojanovic, A.; Carreira, E. M. Angew.
Chem., Int. Ed. 1998, 37, 3124. (c) Baza´n-Tejeda, B.; Bluet, G.; Broustal,
G.; Campagne, J.-M. Chem. Eur. J. 2006, 12, 8358.
Org. Lett., Vol. 9, No. 11, 2007
2177

In conclusion, we report the catalytic desymmetrization
of readily available silylated cyclohexadiene 3 using Cu-
(OTf)2/DIFLUORPHOS as a catalyst system. In the catalytic
desymmetrization reaction, three selectivity problems that
are (1) regioselectivity, (2) diastereoselectivity, and (3)
differentiation of the enantiotopic double bonds were solved.
To our knowledge, this is the first report on a catalytic
desymmetrization reaction in allylsilane chemistry.9 The
cyclohexadienes 2 obtained are highly useful building blocks
for the synthesis of natural products.
Table 3. Reaction of 3 with Various Aromatic Aldehydes
RCHOa
entry
R
dr (syn:anti)
er (syn)
yield (%)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
4-MeC6H4
4-i-PrC6H4
4-t-BuC6H4
4-MeOC6H4
4-FC6H4
14.8:1
14.1:1
24.1:1
28.9:1
7.8:1
4.1:1
4.7:1
9.8:1
17.5:1
8.7:1
6.1:1
7.0:1
2.6:1
3.4:1
10.8:1
2.5:1
6.4:1
96.5:4.5
96:4
97:3
97:3
94:6
60
78
84
59
91
82
85
83
68
55
59
62
71
44
63
44
79
4-ClC6H4
95:5
4-BrC6H4
2-MeC6H4
2-MeOC6H4
3-MeC6H4
3-MeOC6H4
3-BrC6H4
2-furyl
5-Br-2-furyl
2-thienyl
5-Br-2-thienyl
2-naphthyl
92.5:7.5
91.5:8.5
95.5:4.5
96:4
94:6
94:6
94.5:5.5
94:6c
96.5:3.5
95.5:4.5c
93:7
Acknowledgment. We thank the Deutsche Forschungs-
gemeinschaft (International Research Training Group Mu¨n-
ster/Nagoya) for supporting our work. R.U. thanks the
Alexander von Humboldt Foundation for a postdoctoral
fellowship. Solvias AG is acknowledged for donation of
various ligands.
Supporting Information Available: General experimen-
tal procedures, GC and HPLC analysis methods, and
spectroscopic data for new compounds. This material is
available free of charge via the Internet at http://
pubs.acs.org.
a Conditions: THF at -25 °C for 20 h using 3 (3 equiv), Cu(OTf)2 (10
mol %), 6 (10 mol %), CF3CH2OH (3 equiv), TBAF (1 equiv). and RCHO
(1 equiv). b Regioisomer 4 was formed in <2% yield. c Determined after
debromination.
OL070689D
<2% in these reactions, and moderate to excellent yields
were achieved (44-91%). The reactions with para-substi-
tuted aromatic aldehydes occurred with good enantioselec-
tivities and moderate to excellent syn:anti selectivities (entries
1-7); ortho- and meta-substituted aromatic aldehydes and
heteroaromatic aldehydes work equally well (entries 8-17).17
(17) Aliphatic aldehydes delivered worse results. Reaction with cyclo-
hexane carbaldehyde afforded 2 (R ) C6H11) with a 3.7:1 syn:anti selectivity
along with the 1,4-diene 4 (R ) C6H11) (ratio 2 (R ) C6H11):4 (R ) C6H11)
) 2.5:1) in a low combined yield (19%). The ee of the syn,product was
not determined. Pivalaldehyde did not react under the optimized conditions
with the cyclohexadienyl Cu species.
2178
Org. Lett., Vol. 9, No. 11, 2007