Tandem intramolecular silylformylation and silicon-assisted cross-coupling reactions. Synthesis of geometrically defined α,β-unsaturated aldehydes

Article abstract of DOI:10.1021/jo034064e

The palladium- and copper-catalyzed cross-coupling reactions of cyclic silyl ethers with aryl iodides are reported. Silyl ethers 3 were readily prepared by intramolecular silylformylation of homopropargyl silyl ethers 2 under a carbon monoxide atmosphere. The reaction of cyclic silyl ethers 3 with various aryl iodides 7 in the presence of [(allyl)PdCl]₂, CuI, a hydrosilane, and KF·2H₂O in DMF at room temperature provided the α,β-unsaturated aldehyde coupling products 8 in high yields. The need for copper in this process suggested that transmetalation from silicon to copper is an important step in the mechanism. Although siloxane 3 and the product 8 are not stable under basic conditions, KF·2H₂O provided the appropriate balance of reactivity toward silicon and reduced basicity. The addition of a hydrosilane to [(allyl)PdCl]₂ was needed to reduce the palladium(II) to the active palladium(0) form.

Full text of DOI:10.1021/jo034064e

Ta n d em In tr a m olecu la r Silylfor m yla tion a n d Silicon -Assisted
Cr oss-Cou p lin g Rea ction s. Syn th esis of Geom etr ica lly Defin ed
r,â-Un sa tu r a ted Ald eh yd es
Scott E. Denmark* and Tetsuya Kobayashi
Department of Chemistry, Roger Adams Laboratory, University of Illinois, Urbana, Illinois 61801
denmark@scs.uiuc.edu
Received J anuary 20, 2003
The palladium- and copper-catalyzed cross-coupling reactions of cyclic silyl ethers with aryl iodides
are reported. Silyl ethers 3 were readily prepared by intramolecular silylformylation of homopro-
pargyl silyl ethers 2 under a carbon monoxide atmosphere. The reaction of cyclic silyl ethers 3
with various aryl iodides 7 in the presence of [(allyl)PdCl]2, CuI, a hydrosilane, and KF‚2H2O in
DMF at room temperature provided the R,â-unsaturated aldehyde coupling products 8 in high yields.
The need for copper in this process suggested that transmetalation from silicon to copper is an
important step in the mechanism. Although siloxane 3 and the product 8 are not stable under
basic conditions, KF‚2H2O provided the appropriate balance of reactivity toward silicon and reduced
basicity. The addition of a hydrosilane to [(allyl)PdCl]2 was needed to reduce the palladium(II) to
the active palladium(0) form.
SCHEME 1
In tr od u ction a n d Ba ck gr ou n d
The formation of carbon-carbon bonds by palladium-
or nickel-catalyzed cross-coupling reactions of organo-
metallic reagents is one of the most powerful carbon-
carbon bond-forming reactions in organic chemistry.1
Although the Suzuki, Stille, and Negishi couplings are
well-known as very useful reactions, they each have
distinct drawbacks such as the requirement of forcing
conditions, the involvement of toxic reagents and byprod-
ucts, as well as oxygen and moisture sensitivity.1a,2
Recently, the cross-coupling reactions of organosilicon
compounds have been extensively developed as an alter-
native method which can provide practical solutions to
these problems.3
In the past few years, reports from these laboratories
have described new variants of organosilicon cross-
coupling reactions for the stereospecific construction of
bonds between unsaturated carbon functions.4 For ex-
ample, the synthesis of trisubstituted homoallylic alco-
hols by means of either a syn (platinum)5a and an anti
(ruthenium)5b selective hydrosilylation reaction of alkyn-
ylhydrosilanes followed by palladium-catalyzed cross-
coupling has recently been reported (Scheme 1). The
intramolecular hydrosilylation of alkynylsilyl ethers af-
fords cyclic siloxanes having a stereodefined alkenyl
moiety. Thus, subsequent stereospecific cross-coupling
(1) (a) Metal-catalyzed Cross-coupling Reactions; Diederich, F.,
Stang, P. J ., Eds.; Wiley-VCH: Weinheim, Germany, 1998. (b) Heck,
R. F. Palladium Reagents in Organic Syntheses; Academic Press:
London, 1985. (c) Tsuji, J . Palladium Reagents and Catalysts. Innova-
tions in Organic Synthesis; Wiley: Chichester, UK, 1995.
(2) (a) Suzuki, A. In Handbook of Organopalladium Chemistry for
Organic Synthesis; Negishi, E., Ed.; Wiley: Hoboken, NJ , 2002; p 249.
(b) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457. (c) Suzuki, A.
In Metal-catalyzed Cross-coupling Reactions; Diederich, F., Stang, P.
J ., Eds.; Wiley-VCH: Weinheim, Germany, 1998; Chapter 2. (d) Stille,
J . K. Angew Chem., Int. Ed. Engl. 1986, 25, 508. (e) Farina, V.;
Krishnamurthy, V.; Scott, W. J . Org. React. 1998, 50, 1. (f) Mitchell,
T. N. In Metal-catalyzed Cross-coupling Reactions; Diederich, F., Stang,
P. J ., Eds.; Wiley-VCH: Weinheim, Germany, 1998; Chapter 4. (g)
Negishi, E.; Liu, F. In Metal-catalyzed Cross-coupling Reactions;
Diederich, F., Stang, P. J ., Eds.; Wiley-VCH: Weinheim, Germany,
1998; Chapter 1.
(3) (a) Barcza, S. In Silicon Chemistry; Corey, J . Y., Corey, E. R.,
Gaspar, P. P., Eds.; Ellis Harwood: Chichester, UK, 1987. (b) Tamao,
K.; Kobayashi, K.; Ito, Y. Tetrahedron Lett. 1989, 30, 6051. (c)
Hatanaka, Y.; Hiyama, T. Synlett 1991, 845. (d) Hiyama, T. In Metal-
Catalyzed Cross-coupling Reactions; Diederich, F., Stang, P. J ., Eds.;
Wiley-VCH: Weinheim, Germany, 1998; Chapter 10. (e) Mowery, M.
E.; DeShong, P. J . Org. Chem. 1999, 64, 1684. (f) Mowery, M. E.;
DeShong, P. Org. Lett. 1999, 1, 2137. (g) Mowery, M. E.; DeShong, P.
J . Org. Chem. 1999, 64, 3266. (h) Lee, H. M.; Nolan, S. P. Org. Lett.
2000, 2, 2053. (i) Lam, P. Y. S.; Deudon, S.; Averill, K. M.; Li, R.; He,
M. Y.; DeShong, P.; Clark, C. G. J . Am. Chem. Soc. 2000, 122, 7600.
(j) Denmark, S. E.; Sweis, R. F. Chem. Pharm. Bull. 2002, 50, 1531.
(4) (a) Denmark, S. E.; Sweis, R. F. Acc. Chem. Res. 2002, 35, 835.
(b) Denmark, S. E.; Yang, S.-M. J . Am. Chem. Soc. 2002, 124, 2102.
(c) Denmark, S. E.; Sweis, R. F. Org. Lett. 2002, 4, 3771.
(5) (a) Denmark, S. E.; Pan, W. Org. Lett. 2001, 3, 61. (b) Denmark,
S. E.; Pan, W. Org. Lett. 2002, 4, 4163.
10.1021/jo034064e CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/05/2003
J . Org. Chem. 2003, 68, 5153-5159
5153

Denmark and Kobayashi
TABLE 1. Silylfor m yla tion of Alk yn ylh yd r osila n e 2
reactions afford geometrically homogeneous trisubsti-
tuted alkenes. The geometry of the alkene moiety is
predictable and depends on the stereoselectivity of the
hydrosilylation.
Introduction of both a carbon-carbon bond and a
carbon-silicon bond, as for example in intramolecular
silylformylation of an alkynylsilane, is a well-studied and
powerful synthetic method for introduction of an alde-
hyde function into organic molecules with concomitant
formation of a defined alkenyl unit.6 Recently, novel
tandem processes employing the intramolecular silyl-
formylation have been reported.6b,c To expand the scope
of organosilicon couplings, we envisioned the use of
intramolecular silylformylation for the preparation of
unsaturated silicon substrates that could provide access
to a diverse family of R,â-unsaturated aldehydes bearing
hydroxyalkyl groups. Herein, we report the development
and application of a novel method for the preparation of
stereodefined, functionalized R,â-unsaturated aldehydes
in three steps starting from alkynyl alcohols.
substrate
CO,
psi
time,
h
products
(yield, %)
entry
(n)
catalyst
1
2
3
4
5
2a (1)
2a (1)
2b (1)
2c (2)
2c (2)
5a
6b
6
6
6
150
150
150
150
750
9
9
9
20
20
3a (26)
3a (72)
3b (72)
3c (8) + 4c
3c (18) + 4c (2)
c
Rh(acac)(CO)2. [Rh(CNt-Bu)4][Co(CO)4]. c Observed on TLC
but not isolated.
a
b
with the bimetallic catalyst 6 also proceeded smoothly
to give the cyclic silyl ether 3b bearing a methyl group
on the alkene in 72% yield (entry 3). However, under
similar reaction conditions, the reaction of 2c gave only
8% of the six-membered cyclic silyl ether 3c. Employing
higher pressures (i.e., 750 psi) led to no improvement in
the yield (entries 4 and 5). The geometries of the double
bonds in 3b and 3c were also established by NOE
experiments.
2. Cr oss-Cou p lin g Rea ction of th e Cyclic Silyl
Eth er . Initial studies on the cross-coupling reactions of
3a revealed that it possessed reactivity distinct from that
of other alkenylsilanes. The reaction conditions commonly
used for organosilicon cross-coupling reactions (Pd2(dba)3‚
CHCl3, TBAF, THF) failed, presumably because the silyl
ether is unstable. For example, the reaction of cyclic silyl
ether 3a with 4-iodoacetophenone (7a ) in the presence
of Pd2(dba)3‚CHCl3 and TBAF in THF did not yield the
desired product (Table 2, entry 1) but rather the homo-
coupling product, 9a .8 Formation of the homocoupling
Resu lts
1. In tr a m olecu la r Silylfor m yla tion of Alk yn yl
Hyd r osila n es. The substrates for intramolecular silyl-
formylation, alkynyloxy hydrosilanes 2, were easily
prepared in excellent yield from alkynyl alcohols 1
(Scheme 2). The intramolecular silylformylation was
initially carried out following a procedure described by
Ojima.7
SCHEME 2
TABLE 2. Cr oss-Cou p lin g Rea ction of 3a
Two catalysts, Rh(acac)(CO)2 (5) and [Rh(CNt-Bu)4][Co-
(CO)4] (6), were tested for the silylformylation of 2a .
These rhodium complexes are reported to be effective
catalysts that can provide aldehydes in high yield.7 The
intramolecular silylformylation of 2a was performed in
the presence of a rhodium catalyst under a carbon
monoxide atmosphere (150 psi) in toluene at 70 °C.
Although the commercially available rhodium catalyst
5 gave the cyclic silyl ether 3a only in 26% yield, the
rhodium-cobalt catalyst 6 afforded the same product in
72% yield (Table 1, entries 1 and 2). The 1H NMR
spectrum of 3a displayed a doublet at δ 9.46 ppm for the
aldehyde group and doublet of triplets at δ 6.82 ppm for
the olefinic proton. The olefin geometry in 3a was
determined by a NOE experiment. A measurable (5%)
NOE was observed between the olefinic proton (δ 6.82
ppm) next to the aldehyde and allylic methylene protons
(δ 2.81 ppm). The reaction of alkynyloxy hydrosilane 2b
products
(yield,a %)
entry
conditions
8a 9a 10a
1
2
3
4
Pd2(dba)3‚CHCl3 TBAF, THF, rt, 24 h
Pd2(dba)3‚CHCl3 TBAF, THF, 50 °C, 2 h
Pd2(dba)3‚CHCl3 TBAF, THF, 50 °C, 3 hb
0
0
0
28
46
40
0
0
0
9
APC,c CsF, DMF, CuI (5 mol %), rt, 30 min 55 13
a
b
Yield of isolated product. Aryl iodide was added in three
portions every 20 min. c APC ) [(allyl)PdCl]2.
(6) Recent reviews: (a) Ojima, I.; Li, Z.; Zhu, J . The Chemistry of
Organic Silicon Compounds; Rappoport, Z., Apeloig, Y., Eds.; Wiley:
Chichester, 1998; Vol. 2, Part 2, Chapter 29. See also: (b) O’Malley,
S. J .; Leighton, J . L. Angew. Chem., Int. Ed. 2001, 40, 2915. (c)
Bonafoux, D.; Ojima, I. Org. Lett. 2001, 3, 1303. (d) Zacuto, M. J .;
Leighton, J . L. J . Am. Chem. Soc. 2000, 122, 8587.
(7) (a) Ojima, I.; Vidal, E.; Tzamarioudaki, M.; Matsuda, I. J . Am.
Chem. Soc. 1995, 117, 6797. (b) Monteil, F.; Matsuda, I.; Alper, H. J .
Am. Chem. Soc. 1995, 117, 4419.
product is often observed in transition-metal-catalyzed
cross-coupling reactions.9 Even at elevated temperatures,
(8) Chao, C. S.; Cheng, C. H.; Chang C. T. J . Org. Chem. 1983, 48,
4904.
(9) Denmark, S. E.; Wu, Z. Org. Lett. 1999, 1, 1495 and references
therein.
5154 J . Org. Chem., Vol. 68, No. 13, 2003

Geometrically Defined R,â-Unsaturated Aldehydes
the desired product could not be obtained (entry 2).
Because the maintenance of low aryl iodide concentration
often suppresses a formation of homocoupling products,5a
the aryl iodide 7a was added in three portions every 20
min. Unfortunately, even this modification did not work
and again only homocoupling product 9a was isolated in
40% yield (entry 3). [(Allyl)PdCl]2, which is soluble in
THF (unlike Pd2(dba)3‚CHCl3), was employed as the
palladium catalyst, but the desired product was not
obtained at all. Because the silyl ether seemed to be
unstable to TBAF, a weaker fluoride activator CsF was
employed in DMF (to increase the solubility of CsF), but
again no desired product was observed.
Clearly, the problem with this coupling process was
more fundamental, as variation in the common reaction
components did not give even a hint of the desired
products. We surmised that the electron-deficient nature
of the alkenylsilane unit (the greatest structural change
compared to other substrates) was responsible for the
failure, presumably by retarding the crucial transmeta-
lation step. In recent years, the use of cocatalysts in
palladium-catalyzed cross-coupling reactions has been
effective in improving reaction rates and product yields.10
In particular, copper cocatalysts have been used success-
fully in Stille cross-coupling reactions of electronic-
deficient alkenylstannanes.10a,b Organosilicon cross-
coupling reactions also benefit from the use of copper
cocatalysts.11 Among many readily available copper salts,
CuI is frequently employed as the cocatalyst and often
gives better results than other copper salts.10b,11d,g To our
delight, the use of CuI in the present cross-coupling
reaction showed a remarkable effect. Thus, reaction of
3a with 7a in the presence of both [(allyl)PdCl]2 and CuI
with CsF as the activator in DMF at room temperature
gave the desired product 8a in 55% yield along with
homocoupling product 9a and hemiacetal 10a in 13 and
9% yields, respectively (Table 2, entry 4).
well as the sources of copper and palladium catalysts,
the fluoride activator, addition of water and ligands, and
use of hydrosilane as additives were systematically
surveyed as described below.
2.1. Effect of th e Solven t. To identify a suitable
solvent for the cross-coupling reaction, 1.2 equiv of 3a
with respect to 7a was used at room temperature under
an argon atmosphere. The reaction in DMF afforded a
rapid reaction, and HPLC analysis after 5 min showed
73% of 8a and 15% of 9a . No hemiacetal 10a was
observed (Table 3, entry 1). Although the reaction in
acetonitrile gave the product 8a , the reaction was slower
than that in DMF (entry 2). Use of THF and 1,4-dioxane
afforded little desired product even after 10 h (entries 3
and 4). Thus, DMF was selected as the appropriate
solvent for both rapid conversion and high yield.
TABLE 3. Op tim iza tion of th e Solven t
HPLC yield,b %
time,
entry
1
solvent
DMF
min
7a
8a
9a
10a
5
10
5
60
600
600
4
73
53
3
53
11
1
15
19
5
17
3
<1
10
<1
8
6
<1
2
54
<1
67
73
2
CH3CN
3
4
THF
1,4-dioxane
2
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard.
2.2. Effect of Cu a n d P d Sou r ces a n d Th eir Ra tio.
Next, optimization studies focused on identification of the
appropriate ratio of copper to palladium in DMF as a
solvent (Table 4). The initial reaction conditions that
employed 5 mol % of [(allyl)PdCl]2 and 10 mol % of CuI
(Pd/Cu ) 1.0) proved to be optimal (Table 4, entry 3).
Decreasing in the amount of CuI tended to slow the
reaction (entries 1 and 2). For example, with 5 mol % of
[(allyl)PdCl]2 and 5 mol % of CuI (Pd/Cu ) 2.0), the
reaction was not complete within 30 min. Increasing the
amount of CuI increased the rate of formation of the
homocoupling product 9a (entries 4 and 5). Further
At this point, although the desired product could be
obtained, formation of the homocoupling product 9a and
the hemiacetal 10a remained problematic. To suppress
the formation of these side products and to obtain the
products in higher yields, we undertook further optimiza-
tion studies by HPLC analysis using biphenyl as an
internal standard. The ratio of copper to palladium as
(10) Use of copper cocatalyst in the Stille and Suzuki reactions. (a)
Marino, J . P.; Long, J . K. J . Am. Chem. Soc. 1988, 110, 7916. (b)
Liebeskind, L. S.; Fengl, R. W. J . Org. Chem. 1990, 55, 5359. (c) Farina,
V.; Kapadia, S.; Krishnan, B.; Wang, C.; Liebeskind, L. S. J . Org. Chem.
1994, 59, 5905. (d) Takeda, T.; Kabasawa, Y.; Fujisawa, T. Tetrahedron,
1995, 51, 2515. (e) Han, X.; Stolz, B. M.; Corey, E. J . J . Am. Chem.
Soc. 1999, 121, 7600. (f) Xu, J .; Burton, D. J . Tetrahedron Lett. 2002,
43, 2877. (g) Liu, Q.; Burton, D. J . Org. Lett. 2002, 4, 1483. (h) Liu,
X.-x.; Deng, M.-z. Chem. Commun. 2002, 622.
TABLE 4. Op tim iza tion of th e Cu /P d Ra tio
HPLC yield,b %
catalysts
(mol %)
time,
min
(11) Silicon on an sp2 carbon: (a) Taguchi, H.; Ghoroku, K.; Tadaki,
M.; Tsubouchi, A.; Takeda, T. J . Org. Chem. 2002, 67, 8450. (b)
Taguchi, H.; Miyashita, H.; Tsubouchi, A.; Takeda, T. Chem. Commun.
2002, 2218. (c) Taguchi, H.; Ghoroku, K.; Tadaki, M.; Tsubouchi, A.;
Takeda, T. Org. Lett. 2001, 3, 3811. (d) Hanamoto, T.; Kobayashi, T.;
Kondo, M. Synlett 2001, 281. (e) Ito, H.; Sensui, H.; Arimoto, K.; Miura,
K.; Hosomi, A. Chem. Lett. 1997, 639. (f) Xi, Z.; Fischer, R.; Hara, R.;
Sun, W.-H.; Obora, Y.; Suzuki, N.; Nakajima, K.; Takahashi, T. J . Am.
Chem. Soc. 1997, 119, 12842. (g) Suginome, M.; Kinugasa, H.; Ito Y.
Tetrahedron Lett, 1994, 35, 8635. Silicon on an sp3 carbon. (h) Takeda,
T.; Uruga, T.; Gohroku, K.; Fujiwara, T. Chem. Lett. 1999, 821. (i)
Yoshida, J .; Tamao, K.; Kurita, A.; Kumada, M. Tetrahedron Lett, 1978,
1809. Silicon on an sp carbon. (j) Nishihara, Y.; Takemura, M.; Mori,
A.; Osakada, K. J . Organomet. Chem. 2001, 620, 282. (k) Nishihara,
Y.; Ikegashira, K.; Hirabayashi, K.; Ando, J .; Mori, A.; Hiyama, T. J .
Org. Chem. 2000, 65, 1780.
entry
1
Cu/Pd
0.25
7a
8a
9a
10a
APCa (5)
CuI (2.5)
APC (5)
CuI (5)
APC (5)
CuI (10)
APC (5)
CuI (20)
APC (5)
CuI (40)
5
30
5
30
5
10
5
10
5
53
39
24
18
4
23
23
46
24
73
53
55
56
72
72
8
15
17
23
15
19
23
27
26
25
<1
7
4
27
<1
10
2
8
<1
4
2
3
4
5
0.5
1
2
2
<1
<1
<1
<1
4
10
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard.
J . Org. Chem, Vol. 68, No. 13, 2003 5155

Denmark and Kobayashi
TABLE 5. Su r vey of Cu Coca ta lysts in Com bin a tion
w ith [(a llyl)P d Cl]2
TABLE 6. Su r vey of P d Ca ta lysts in Com bin a tion w ith
Cu I
HPLC yield,b %
time,
HPLC yield,b %
time,
entry
Pd (mol %)
APCa(5)
min
7a
8a
9a
10a
entry
Cu (mol %)
CuI (10)
min
7a
8a
9a
10a
1
2
5
5
10
10
30
5
10
10
30
10
30
5
4
14
5
10
5
73
43
39
50
24
30
25
3
2
4
<1
29
24
15
32
33
12
12
30
35
8
9
3
4
17
18
<1
<1
7
13
30
7
15
7
6
1
2
5
5
4
5
2
73
56
56
9
8
10
3
71
57
49
21
8
5
19
9
15
19
27
17
23
10
10
18
19
16
21
25
42
17
27
12
22
22
36
<1
<1
5
<1
3
12
18
<1
11
6
26
<1
9
PdCl2 (10)
CuBr (10)
10
10
30
10
30
5
10
10
30
10
30
10
30
10
30
10
30
3
4
Pd(PPh3)2Cl2 (10)
Pd(OAc)2 (10)
Pd(dba)2 (10)
Pd(PPh3)4 (10)
Pd(PPh3)4 (10)
3
4
CuCl (10)
51
43
62
51
9
<1
10
<1
40
29
37
18
50
26
38
21
14
8
CuCN (10)
5
60
54
46
43
41
27
5
CuOTf‚toluene (10)
CuCl2 (10)
6
6
6
13
<1
7
7c
7
CuF2 (10)
10
a
b
8
Cu(OTf)2 (10)
Cu(BF4)2‚6H2O (10)
CuSO4‚5H2O (10)
9
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard. c CuCl2 (10 mol %) was used.
14
<1
4
<1
8
9
18
31
32
19
10
4-iodoacetophenone (7a ) in the presence of CsF was
almost complete within 5 min. However, longer reaction
times yielded larger amounts of the rearranged hemiac-
etal compound. Thus, the hemiacetal 10a is apparently
derived from the cross-coupling product 8a (See Discus-
sion). Thus, slower reacting aryl halides would likely lead
to a greater proportion of this side reaction. Accordingly,
the search for a milder activator became a priority.
Potassium fluoride also has been used as an activator
for organosilicon cross-coupling reactions.12 Thus, com-
mercially available, spray-dried KF was employed (Table
7, entry 2). Although the reaction required a considerably
longer time (>15 h) as compared to CsF, HPLC analysis
showed a higher yield of the coupling product 8a and
smaller amounts of the homocoupling product 9a .
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard.
optimization focused on identifying superior sources of
the copper and palladium catalysts. In the absence of an
understanding of the reaction mechanism, it was unclear
which oxidation states of the copper and palladium
agents were active in the carbon-carbon bond-formatting
process. A survey of copper and palladium sources was
undertaken that maintained a 1/1 ratio of the two metals
as was shown above.
Other Cu(I) cocatalysts in combination with [(allyl)-
PdCl]2 were generally inferior to CuI (Table 5, entries
2-4). However, Cu(I)OTf-toluene complex gave results
comparable to CuI (entry 5). Interestingly, Cu(II) cata-
lysts such as CuCl2, Cu(OTf)2, Cu(BF4)2, and CuSO4 also
gave the desired coupling product, but in each case,
HPLC analysis showed a significant amount of the
homocoupling product as well (entries 6-10). Thus, CuI
seemed to be the best cocatalyst in combination with
[(allyl)PdCl]2.
Several palladium sources were surveyed in the pres-
ence of CuI (10 mol %) as the cocatalyst (Table 6).
Palladium(II) catalysts such as PdCl2, Pd(PPh3)2Cl2, and
Pd(OAc)2 were tested, but all these Pd(II) catalysts were
inferior to [(allyl)PdCl]2 (Table 6, entries 1-4). Next, a
survey of Pd(0) sources was conducted. Most palladium
cross-coupling reactions proceed through a Pd(0) to Pd-
(II) cycle because Pd(0) is required for the oxidative
addition step.1 Hence, Pd(0) sources may allow for faster
and more efficient reactions. Surprisingly, the use of a
Pd(0) catalyst such as Pd(dba)2 or Pd(PPh3)4 with CuI
slowly afforded the product 8a in poor yield (entries 5
and 6) as did the coupling reaction in the presence of Pd-
(dba)2 and CuCl2 (entry 7). As a result, it was concluded
that the combination of [(allyl)PdCl]2 and CuI was
optimal (Table 6, entry 1).
TABLE 7. Su r vey of F lu or id e Sou r ces for
Cr oss-Cou p lin g
HPLC yield,b %
3a
time,
min
entry
(equiv)
activator
7a
8a
9a
10a
1
2
1.2
1.2
CsF
KFc
5
300
900
300
600
60
4
73
57
78
81
80
71
90
15
7
10
9
9
7
<1
<1
<1
<1
3
<1
<1
33
14
6
5
19
<1
3
4
1.2
1.3
KF‚2H2O
KF‚2H2O
120
9
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard. c Spray-dried KF purchased from
Aldrich was used.
Previous studies of these laboratories have shown that
water strongly effects the nucleophilicity and/or basicity
of the fluoride activator to reduce the formation of
2.3. Su r vey of F lu or id e Sou r ce. As shown in Table
4, entry 3, the reaction of cyclic silyl ether 3a and
(12) Hatanaka, Y.; Goda, K.; Okahara, Y.; Hiyama, T. Tetrahedron
1994, 50, 8301.
5156 J . Org. Chem., Vol. 68, No. 13, 2003

Geometrically Defined R,â-Unsaturated Aldehydes
byproducts.4c Thus, commercially available KF‚2H2O was
tested as an activator. The reaction with KF‚2H2O
proceeded at a considerably faster rate; after 5 h, 81% of
8a and 9% of 9a were observed by HPLC (Table 7, entry
3). However, the reaction was not complete even after
10 h. For the complete consumption of iodide, 1.3 equiv
of 3a was used. This additional 0.1 equiv of 3a had a
dramatic effect. The reaction was complete within 2 h,
and HPLC analysis showed the formation of 8a and 9a
in 90 and 9% yields, respectively (entry 4). It is notewor-
thy that even after an extended reaction time, almost no
hemiacetal product 10a was observed.
The obvious dependence on water content in the
fluoride source motivated a search for the appropriate
amount of water in this reaction. In these experiments,
water was added to a suspension of anhydrous KF in
DMF. Although the reaction rates relative to anhydrous
KF increased by the addition of water (up to 40 equiv),
the reactions were still slower than with KF‚2H2O (Table
8). High rates and yields are specific to use of KF‚2H2O,
presumably because of its solubility in DMF. From these
studies, it is clear that KF‚2H2O is the most suitable
activator for suppressing hemiacetal formation and de-
creasing the amount of the homocoupling product.
TABLE 9. In Situ Red u ction of [(a llyl)P d Cl]2 w ith
Va r iou s Hyd r osila n es
HPLC yield,b %
source
entry of 3a
time,
h
additive
7a 8a 9a 10a
1
2
c
d
none
none
2
2
20
2
2
2
2
2
1
2
<1
49
9
1.5 87
<1
<1
1.5 89
1.8 90
2
10
3
90
26
9
6
<1
<1
72 14 <1
3
4
5
6
7
8
9
d
d
d
d
d
d
d
(i-Pr2SiH)2O (11)
i-Pr2Si(H)OH (12)
[-MeSi(H)O-]4 (13)
[-MeSi(H)O-]3-5 (14)
Et3SiH (15)
9
9
9
9
8
<1
<1
<1
<1
<1
89
89
PhSi(Me)2H (16)
EtOSi(Me)2H (17)
85 11 <1
79
84
7
8
<1
<1
5
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard. c Distilled material. Chromato-
d
graphed material.
However, simple fractional distillation did not provide
analytically pure materials. The 1H NMR analysis of the
distilled material indicated approximately 95% purity,
which was considered to be sufficient for use in subse-
quent cross-coupling reactions.
TABLE 8. Effect of th e Am ou n t of Wa ter
Although silica gel chromatography did cause the
decomposition of the silyl ether (ca. 30-80% recovery),
small amounts of pure 3a could be obtained. Surprisingly,
coupling reactions that employed chromatographed ma-
terial proved to be considerably slower under the optimal
conditions (Table 9, entry 2). This prompted further
investigation of the trace impurities present in 3a and
their possible role in the catalytic cycle. However, further
purification of the distillated material by chromatography
was fairly difficult. Whereas aluminum oxide destroyed
3a , it was found that 3a was stable to Florisil. Chro-
matographic purification using Florisil and silica gel gave
pure material without decomposition.13 Two side prod-
ucts, (i-Pr2SiH)2O (11) and an unknown hydrosilane were
also found in some fractions.14 Both hydrosilanes could
HPLC yield,b %
water
(equiv)
time,
h
entry
7a
8a
9a
10a
1
2
KF‚2H2O
2
2
10
2
10
2
10
2
10
2
10
2
10
2
10
<1
83
74
92
75
69
46
37
15
34
10
20
6
90
2
12
2
9
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
5
0
<1
<1
<1
<1
<1
7
2
10
4
10
6
9
6
9
3
4
5
6
7
8
2
4
12
10
29
38
67
47
78
62
74
53
72
10
20
40
110
1
be seen in the H NMR spectra of the distilled material
28
7
as trace impurities. These impurities could not be re-
moved from 3a by repeated or fractional distillation.
However, use of the Florisil/silica gel chromatography did
remove them. Apparently, the presence of these hydrosi-
lanes was essential for initiating the catalytic cycle. This
was proven by addition of 0.025 equiv of 11 to a mixture
of pure 3a and 7a (under standard conditions), which led
to a rapid reaction (Table 9, entry 3). HPLC analysis
showed almost identical results compared to the reaction
of 3a , which contained hydrosilane impurities. The
addition of 11 to the reaction mixture caused immediate
formation of a black precipitate, presumably colloidal Pd-
(0). The precipitation was also observed in the reaction
of 3a contaminated with hydrosilane impurities.
a
b
APC ) [(allyl)PdCl]2. Determined by HPLC analysis using
biphenyl as an internal standard.
2.4. Su r vey of Liga n d s. Having addressed the prob-
lems of reaction rate (and indirectly hemiacetal byproduct
formation), we turned our attention to the problem of
homocoupling. Previous studies have revealed that the
formation of homocoupling products can often be sup-
pressed by addition of a phosphine or an arsine ligand.9
The effect of the ligand was also tested in reactions
employing KF as an activator. Chelating phosphines such
as DPPF and DPPP, as well as triphenylphosphine and
triphenylarsine, were tested, and in no case was any
significant improvement observed (6-10% of 9a ).
Several hydrosilanes were next examined as additives
(Table 9, entries 4-9). These reagents accelerated the
2.5. In Situ Red u ction of [(Allyl)P d Cl]2. A major
concern at the outset of these studies was the purity of
the cyclic silyl ethers 3. Since 3 could not be purified by
silica gel chromatography, distillation was employed.
(13) Chromatography on Florisil did not give good separation.
(14) Tetraisopropyldisiloxane (11) and an unknown hydrosilane were
obtained from the distilled material in about 1 and 3 w/w%, respec-
tively.
J . Org. Chem, Vol. 68, No. 13, 2003 5157

Denmark and Kobayashi
TABLE 10. Cr oss-Cou p lin g Rea ction w ith Va r iou s
Ar om a tic Iod id esa
reaction as did the hydrosilane 11 (entry 3). Among the
hydrosilanes tested, methylhydrocyclosiloxane (14) was
chosen as an additive because of its low cost and ready
availability. The commercially available material is a
mixture of three cyclosiloxanes {[-MeSi(H)O]3-5} and is
easy to handle.
The foregoing extensive optimization revealed that use
of [(allyl)PdCl]2 and CuI in a 1 to 1 ratio is the optimal
combination of catalysts. Although TBAF is commonly
employed for organosilicon cross-coupling reactions, in
this case, KF‚2H2O is a suitable activator in terms of the
yield, reaction rate, and minimization of byproduct.
Finally, in situ reduction of [(allyl)PdCl]2 by the hydrosi-
lane was necessary for rapid reaction.
yield,c%
silyl
entry ether
time,
h
aryl iodide
8
9
1
2
3
4
5
6
7d
8
9
3a
3a
3a
3a
3a
3a
3a
3a
3a
4-(CH3CO)C6H4I (7a )
4-(EtO2C)C6H4I (7b)
4-O2NC6H4I (7c)
C6H5I (7d )
4-MeOC6H4I (7e)
2-MeC6H4I (7f)
2
2
1
2
2
2
4
3
2
91
92
83
93
87
57
79
78
88
7
5
13
(4)f
5
(11)f
(18)f
(7)f
(8)f
3. Su r vey of Electr op h iles in th e Cr oss-Cou p lin g
Rea ction . With optimal reaction conditions in hand,
several electrophiles were examined. The reaction of 3a
with 4-iodoacetophenone (7a ) under the optimal condi-
tions proceeded smoothly to completion in 2 h. The
desired product 8a was isolated in 91% yield along with
7% of homocoupling product 9a (Table 10, entry 1). The
1H NMR spectrum of 8a showed a doublet at δ 9.43 ppm
for the aldehyde and two broad doublets at δ 8.01 and
7.42 ppm for the aromatic protons. The olefin geometry
was determined as before by NOE experiments. Reaction
of 3a with ethyl 4-iodobenzoate (7b) also proceeded
cleanly, and the ester functionality was compatible under
these conditions. The desired product 8b could be ob-
tained in 92% yield (entry 2). 4-Iodonitrobenzene (7c)
reacted with 3a rapidly, and the reaction was complete
within 1 h to afford 8c in 83% yield (entry 3). In this
case, however, a larger amount (13%) of homocoupling
product 9c was also obtained. Iodobenzene (7d ) gave a
cleaner reaction, and the product 8d was isolated in 93%
yield (entry 4). Electrophiles such as 4-iodoanisole (7e)
bearing electron-donating groups often react more slowly
than those bearing electron-withdrawing groups. Under
the optimal conditions, the reaction of 7e with 3a was
also complete within 2 h to give 8e in 87% yield (entry
5). Steric hindrance near the coupling center such as in
2-iodotoluene (7f) gave the product 8f in 57% yield; even
after extended time, the reaction did not go to completion.
However, the use of 1.5 equiv of 3a and 4.0 equiv of
KF‚2H2O gave 8f in 79% yield (entries 6 and 7). Although
the reaction with ethyl 2-iodobenzoate was also tested,
1H NMR analysis showed only small amounts of the
product (∼5%). Heteroaromatic iodides 7g and 7h also
gave the products 8g and 8h in 78 and 88% yields,
respectively (entries 8 and 9).
2-MeC6H4I (7f)
2-iodothiophene (7g)
5-iodo-1-phenylsulfonyl-
indole (7h )
4-(CH3CO)C6H4I (7a )
4-(CH3CO)C6H4I (7a )
10
3b
3b
24
7
(36)f (10)f
83
(14)f
11e
a
Reaction was conducted using silyl ether (1.3 equiv), [(allyl)-
PdCl]2 (5 mol %), CuI (10 mol %), and [-MeSi(H)O-]3-5 (2.5 mol
%) unless otherwise mentioned. APC ) [(allyl)PdCl]2. c Yield of
b
d
isolated purified material. Silyl ether (1.5 equiv) and KF‚2H2O
(4.0 equiv) were added. e [(Allyl)PdCl]2 (10 mol %), CuI (20 mol
%), and [-MeSi(H)O-]3-5 (5 mol %) were employed. f Expected
yield from 1H NMR analysis of crude mixture.
Discu ssion
The intramolecular silylformylation of alkynes 2a and
2b proceeded as expected from the literature precedent
and occasions no special comment. However, unlike the
five-membered ring siloxanes prepared previously by
intramolecular hydrosilylation, compounds 3a and 3b
behaved very differently in the cross-coupling process.
Clearly, the electron-withdrawing formyl group greatly
attenuated the reactivity of the siloxane under the
conventional reaction conditions. Fortunately, the addi-
tion of catalytic amounts of copper iodide reinstated the
desired reactivity and cross-coupling could then proceed
rapidly at room temperature with a range of aryl iodides.
The beneficial effect of copper in cross-coupling reactions
is precedented,11 and there are some reports suggesting
that transmetalation from silicon to copper may be the
source of the observed effects.11a-d,15 The mechanism of
transmetalation from silicon to palladium in any cross-
coupling process has not been elucidated, but may
reasonably be viewed as involving the silicon moiety as
a nucleophile and the palladium moiety as an electro-
phile. Thus, electron-deficient silanes are expected to
react very slowly, if at all. The use of a shuttle such as
CuI that behaves as a strong alkenophile and still
maintains nucleophilic reactivity toward palladium could
mediate the formation of the diorganopalladium complex.
In organosilicon cross-couplings, TBAF is frequently
employed for activation of the silicon-carbon bond.3 In
initial attempts for the cross-coupling reaction of 3a with
7a , TBAF was not suitable, presumably due to its high
basicity. Further experiments revealed that the cross-
coupling product 8a decomposed in the presence of TBAF.
The reaction of cyclic silyl ether 3b (bearing a methyl
group on the alkene) with 4-iodoacetophenone (7a ) was
slower than the reaction of 3a with 7a . Even after 24 h,
the reaction did not go to completion under the optimal
conditions. However, higher catalyst and hydrosilane
loadings gave the desired product 8i in 83% yield, after
7 h (entry 11). The geometries of the olefinic moiety of
all the products 8b-i were also established by NOE
experiments (see Supporting Information).
Finally, the reaction of the six-membered cyclic silyl
ether 3c with 7a was also investigated. Although several
conditions were examined, the coupling product could not
be obtained.
(15) (a) Trost, B. M.; Ball, Z. T.; J oge, T. J . Am. Chem. Soc. 2002,
124, 7922. (b) Nishihara, Y.; Ikegashira, K.; Toriyama, F.; Mori, A.;
Hiyama, T.; Bull Chem. Soc. J pn. 2000, 73, 985. See also ref 8 and
references therein.
5158 J . Org. Chem., Vol. 68, No. 13, 2003

Geometrically Defined R,â-Unsaturated Aldehydes
SCHEME 3
in situ reduction of [(allyl)PdCl]2 is necessary for the
rapid reaction and high yields. Interestingly, the in situ
reduction of [(allyl)PdCl]2 seemed to be the most effective
way to generate Pd(0) for the reaction, as external Pd(0)
sources failed to promote the reaction.
Although, the cross-coupling reactions of 3a and 3b
with various aromatic iodides gave products 8 in excellent
yield, the reaction of 3c with 7a did not proceed at all.
Apparently, the conversion of the silicon unit to a reactive
state for transmetalation under these conditions requires
the added assistance of ring strain. The nature of that
intermediate and the involvement of copper species is
under active study.
Con clu sion
The cross-coupling reaction of cyclic siloxanes derived
from a silylformylation reaction has been demonstrated.
The intramolecular silylformylation was found to be an
operationally simple, high-yielding, and highly reproduc-
ible process for the preparation of cyclic silyl ethers.
Because the intramolecular silylformylation proceeds in
a stereoselective manner, the cyclic silyl ethers obtained
possess a stereodefined olefin moiety. Extensive optimi-
zation of the cross-coupling reaction identified conditions
to make the reaction efficient and high yielding. The
addition of CuI as a cocatalyst was crucial. Although the
role of the copper is not clear, the mechanism might
involve transmetalation from silicon to copper prior to
transmetalation to palladium. Also, this study showed
that KF‚2H2O is an effective activator for the base-
sensitive organosilane substrate and the product. Finally,
it was found that the in situ reduction of [(allyl)PdCl]2
by a hydrosilane is critical for rapid completion of the
desired coupling process. The reaction of cyclic silyl ethers
such as 3a and 3b with various aromatic iodides gave
the cross-coupling products in good to excellent yield.
Mechanistic studies on the role of the copper cocatalyst
and further application of the present reaction conditions
to a more diverse group of coupling partners are currently
under investigation.
Accordingly, the use of a milder activator was required.
Although the reaction employing CsF did afford the cross-
coupling product 8a , hemiacetal 10a was also obtained
from the reaction via the coupling product 8a . Indeed,
treatment of 8a with 2.0 equiv of CsF yielded hemiacetal
10a in 80% yield. The mechanism of hemiacetal forma-
tion may involve deprotonation by CsF at the γ-position
of 8a followed by tautomerization and pyran ring forma-
tion (Scheme 3).
To suppress the formation of hemiacetal 10a , the use
of a much milder activator that possessed weaker basicity
yet maintained sufficient affinity for the silicon atom was
sought. It was found that KF‚2H2O fulfilled these re-
quirements and afforded a clean and rapid reaction. The
addition of water to the reaction mixture employing
anhydrous KF revealed that higher reaction rates and
yields were unique to the preformed dihydrate. This
hydration state helped solubilize KF in DMF without
attenuating its silicophilicity.
Another interesting discovery was the need for trace
amounts of a hydrosilane to initiate and promote comple-
tion of the reaction. Hydrosilane 11, which was isolated
as an impurity from 3a , was probably formed from
diisopropylchlorohydrosilane used in the silylation of
3-butyn-1-ol by hydrolysis followed by dehydration. The
role of the hydrosilane in this reaction is believed to be
reduction of the Pd(II) source to Pd(0).16 Indeed, addition
of the hydrosilanes such as 11-17 to a solution of [(allyl)-
PdCl]2 in DMF immediately formed a black precipitate.
Hence, it is believed that the addition of hydrosilane for
Ack n ow led gm en t. We are grateful to the National
Institutes of Health for generous financial support (R01
GM63167).
Su p p or tin g In for m a tion Ava ila ble: Procedures for the
preparation and characterization of cyclic silyl ethers and all
new coupling products, as well as representative procedures
for coupling reactions. This material is available free of charge
via the Internet at http://pubs.acs.org.
(16) Boukherroub, R.; Chatgilialoglu, C.; Manuel, G. Organometal-
lics 1996, 15, 1508.
J O034064E
J . Org. Chem, Vol. 68, No. 13, 2003 5159