Palladium-catalyzed allylic C-OH functionalization for efficient synthesis of functionalized allylsilanes

Article abstract of DOI:10.1021/ja1096732

A new method is described for palladium-catalyzed allylic silylation using allylic alcohols and disilanes as precursors. The reactions proceed smoothly under mild and neutral conditions, and this method is suitable for synthesis of regioand stereodefined allylsilanes. The presented silylation reaction can be easily extended to include synthesis of allylboronates by change of the dimetallic reagent. The presented synthetic procedure offers a broad platform for the selective synthesis of functionalized allyl metal reagents, which are useful precursors in advanced organic chemistry and natural product synthesis.

Full text of DOI:10.1021/ja1096732

Published on Web 12/16/2010
Palladium-Catalyzed Allylic C-OH Functionalization for Efficient Synthesis of
Functionalized Allylsilanes
Nicklas Selander, Jennifer R. Paasch, and Ka´lma´n J. Szabo´*
Department of Organic Chemistry, Stockholm UniVersity, SE-106 91 Stockholm, Sweden
Received October 27, 2010; E-mail: kalman@organ.su.se
Abstract: A new method is described for palladium-catalyzed
allylic silylation using allylic alcohols and disilanes as precursors.
The reactions proceed smoothly under mild and neutral
conditions, and this method is suitable for synthesis of regio-
and stereodefined allylsilanes. The presented silylation reaction
can be easily extended to include synthesis of allylboronates
by change of the dimetallic reagent. The presented synthetic
procedure offers a broad platform for the selective synthesis
of functionalized allyl metal reagents, which are useful precur-
sors in advanced organic chemistry and natural product
synthesis.
Figure 1. Palladium-catalyzed direct silylation and borylation of allylic
alcohols.
mild and neutral reaction conditions many functional groups, such
as aromatic OH, NO2, OR, and COOMe substituents, are tolerated
in the silylation reactions. The reactions proceeded very cleanly,
as in most cases the desired allylsilane was the only product formed
with a full conversion of the applied allylic alcohol. The reaction
time required for the full conversion of the substrate was longer in
the presence of electron-withdrawing groups (such as 1c) compared
to electron-donating substituents (1a, 1d-e); cf. entries 6 and 1, 7,
8. Primary (such as 1a, 1c-e, 1j), secondary (1b, 1f-g, 1k), tertiary
(1h-i), and cyclic (1j, k) alcohols could easily be converted to
the corresponding allylsilanes.
We have found that not only 2a but also its diphenyl analog 2b
could also be used as a silyl source (entries 2, 10, 12). This is
particularly advantageous when the low boiling point of the desired
trimethylsilanes (TMS) encumbers the isolation of the product, as
in the case of the TMS analog of 4g or 4h. In addition, variation
of the silyl group offers several synthetic advantages, when
allylsilanes are applied as substrates.2 The reaction is also very
robust and easily scalable without further optimization of the
reaction conditions. For example, a 20 times scale up to 3 mmol
of 1a could be carried out without a significant change of the
isolated yield (entry 1).
The regio- and stereoselectivity of the reaction is very high. The
regioselectivity is excellent as the linear allylic product is formed
exclusively. For example, isomeric allylic alcohols 1a and 1b gave
only the linear allylsilane 4a (cf. entries 1 and 5) indicating a
possible allyl-palladium mechanism for the transformation. The
double bond geometry was selectively trans in all acyclic allylsilane
products. Even by starting with cis-alcohol 1j we obtained trans-
allylsilane 4j in high selectivity, which is also a typical signature
of the allyl-palladium mechanism. Starting from stereodefined allylic
alcohol 1k, in which the leaving hydroxy group and the COOMe
substituent are in a cis relationship, the major product is 4k with a
trans configuration of the silyl and COOMe groups. Although the
diastereoselectivity is not completely clean (the trans/cis ratio is
3:1), the major formation of the inversion product 4k suggests that
the silyl group is coordinated to palladium prior to a nucleophilic
attack.12,13 As mentioned above additives were not required to
obtain a high activity and selectivity using our catalytic system. In
fact additives, which are able to coordinate strongly to palladium,
inhibited the reaction. For example, the addition of only 10 mol %
of LiCl to our catalyst system (at 50 °C for 15 h) led to an
incomplete reaction for silylation of 1a and the isolated yield was
Allylsilanes encompass the reactivity of both alkenes and
metal-allyl reagents, which renders these compounds as one of
the most important building blocks in modern organic synthesis.1,2
The silyl group in allylsilane reagents has a hyperconjugative3
directing effect (ꢀ-silyl effect) enabling a highly selective
transformation of the sp2 carbons. The versatile scope of the
reactions includes allylations, such as the Hosomi-Sakurai
reaction;4,5 transition-metal-catalyzed processes, such as the
Hiyama coupling;6,7 and even selective synthesis of function-
alized allylic fluorides.8,9 The widespread use of these useful
synthetic applications demands development of new effective
synthetic methods to obtain regio- and stereodefined allylsilanes.
In this respect transition-metal-catalyzed processes proved to
be the most successful procedures.9-24 In particular, palladium-
catalyzed synthesis of allylsilanes by substitution of allylic
esters12,13,23,25 and ethers14 using disilanes26 as a silyl source
proved to be very useful, because of the relatively easy access
to the precursors and the broad synthetic scope of the reactions.
The cost efficiency and the environmental benefits of these
processes could be substantially increased, if allylic alcohols
(1) were used directly as substrates27,28 without using any
additives. This is a highly challenging task as the hydroxy group
is one of the most reluctant leaving groups, which usually has
to be activated in substitution reactions.
Our group has successfully developed several related catalytic
methods for the use of allylic alcohols as precursors for the synthesis
of allylboronates.29-37 We have now devised a new, inexpensive,
and mild procedure for the preparation of functionalized allylsilanes
from allylic alcohols (Figure 1). Our best conditions (Table 1)
involve the use of hexamethyldisilane (2a) as a silyl source and
Pd(BF4)2(MeCN)4 3 as a catalyst (5 mol %) in a DMSO/MeOH
1:1 mixture for smooth conversion of various allylic alcohols (1)
to allylsilanes (4). The reactions were conducted under mild
conditions (typically at 50 °C) without an inert atmosphere and
without addition of any acids, chloride/acetate salts, or other
additives, which are commonly used in analog palladium-catalyzed
substitution reactions of allylic esters and ethers.12-14 Due to the
9
10.1021/ja1096732  2011 American Chemical Society
J. AM. CHEM. SOC. 2011, 133, 409–411 409

C O M M U N I C A T I O N S
Table 1. Palladium-Catalyzed Allylic C-OH Functionalization for
Pd(OCOCF3)2 proceeded with a similar yield (70%) as with 3 (84%)
under identical conditions (entry 1). However, the yield was
decreased when PdCl2 was used as the catalyst (50%), and no
silylated product 4a was formed at all using Pd(OAc)2 as the catalyst
source. Surprisingly, Pd2(dba)3, which was efficiently used as the
catalyst in the silylation of allylic esters,12,13,23 was completely
inefficient for the silylation of allylic alcohols. The application of
a DMSO/MeOH mixture was also important to obtain a fast and
clean silylation reaction. In pure DMSO the conversion of 1 to 4
is extremely slow, while in pure methanol byproducts appear in
the reaction mixture. Application of DMSO is probably important
to keep the palladium, in particular Pd(0), species in solution during
the catalysis. The reaction can also be conducted in chloroform;
however the yield of the transformation drops because of the
formation of byproducts. For example the isolated yield for the
silylation of 1a decreased from 84% to 55% when the DMSO/
MeOH mixture (entry 1) was replaced by chloroform.
the Synthesis of Organosilanes and Boronatesa
To our delight, a very fast and very selective borylation reaction
occurred when disilane 2 was replaced with bis(pinacolato)diboron
(5) (entries 3, 4, 11, 16) using otherwise identical conditions. It
was found that the borylation was much faster than the silylation
reaction. Thus, the silylation of 1a had to be conducted for 15 h at
50 °C to obtain a complete conversion, while formation of
allylboronate 6a required only 15 min under the same reaction
conditions (cf. entries 1 and 3). In fact the C-OH borylation of 1a
could be completed under neutral conditions in 90 min at room
temperature (entry 4). Likewise, alkyl substituted allylic alcohol
1g reacted rapidly under mild conditions affording linear allylic
boronate 6b. The stereoselectivity of the borylation of 1k (trans/
cis 5:1) is somewhat higher than that of the silylation reaction
providing the inversion product 6c as the major diastereomer. When
silaborane reagent PhMe2Si-Bpin was used in place of 2 or 5 under
the standard conditions, an intractable mixture was obtained along
with rapid precipitation of palladium-black.
In order to obtain more insights into the mechanism of the
reactions we compared the rate of formation of the organometallic
products as a function of the substituent effects in the allylic
precursor, the nature of the leaving group, and the source of the
organometallic functionality (see Supporting Information for the
diagram). One of the most interesting findings is that, under identical
reaction conditions, cinnamyl alcohol 1a is silylated faster than
cinnamyl acetate 8. In the presence of a nitro group in the para
position of the phenyl ring (1c), the rate of silylation is slower than
that for the parent compound 1a, while with a methoxy substituent
the rate is practically unchanged. On the other hand the borylation
reaction of 1a is much faster (about 10 times) than the correspond-
ing silylation reaction.
The mechanism of the above rapid and selective silylation of
allylic alcohols has not been understood in every detail; however
on the basis of the above studies (Table 1) at least a plausible
catalytic cycle can be constructed (Figure 2). Our assumption is
that the catalytic process starts from a Pd(0) species, which may
be generated from 3 by reduction by either 2 or methanol under
our standard conditions. In either case, the formation of a “naked”
Pd(0) species is supposed to form in which the ligands are very
loosely coordinated. The activation of the hydroxy group is probably
the most intriguing reaction step of the catalytic cycle. Although,
allylic alcohols are employed in several transition-metal-catalyzed
processes,27,28 very few reactions proceed under mild conditions,30,38
without external activators, as in the above presented process.
Possibly, the high oxophilicity of silicon leads to the formation of
ate-complex 10. Structurally related hypervalent silylsilicate species
have been previously reported in the literature.39,40 Formation of
a Unless otherwise stated allylic alcohol 1 (0.15 mmol), dimetal
reagent 2 or 5 (0.18 mmol), and catalyst 3 (5 mol %) were dissolved in
a DMSO (0.2 mL)/MeOH (0.2 mL) mixture and stirred for the indicated
times and temperatures. b Reaction conditions: temperature/time ) °C/h.
c Isolated yield (%). d The reaction is scaled up to 3.0 mmol of 1a.
e Product isolated in a 3:1 trans/cis mixture. f Product isolated in a 5:1
trans/cis mixture.
decreased from 84% to 35%. It is even more interesting that addition
of LiOAc completely inhibited the process. We could not find any
trace of 4a in the reaction mixture, when 1a was reacted under our
usual conditions (entry 1) in the presence of 10 mol % LiOAc.
With PPh3 (10 mol %) a very complex reaction mixture was
obtained with low conversion of 1a. The silylation of 1a with
9
410 J. AM. CHEM. SOC. VOL. 133, NO. 3, 2011

C O M M U N I C A T I O N S
Acknowledgment. The authors thank the financial support of
the Swedish Research Council (VR). An ERASMUS stipend by
the EU for J.R.P. is greatly acknowledged.
Supporting Information Available: Detailed experimental proce-
1
dures; characterization and H and 13C NMR spectra of the products.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Brook, M. A. Silicon in Organic, Organometallic, and Polymer Chemistry;
Wiley: Chichester, 2000.
(2) Fleming, I.; Barbero, A.; Walter, D. Chem. ReV. 1997, 97, 2063.
(3) Wierschke, S. G.; Chandrasekhar, J.; Jorgensen, W. L. J. Am. Chem. Soc.
1985, 107, 1496.
(4) Hosomi, A.; Miura, K. Bull. Chem. Soc. Jpn. 2004, 77, 835.
(5) Hosomi, A.; Shirahata, A.; Sakurai, H. Tetrahedron Lett. 1978, 33, 3043.
(6) Denmark, S. E.; Werner, N. S. J. Am. Chem. Soc. 2008, 130, 16382.
(7) Hatanaka, Y.; Ebina, Y.; Hiyama, T. J. Am. Chem. Soc. 1991, 113, 7075.
(8) Hazari, A.; Gouverneur, V.; Brown, J. M. Angew. Chem., Int. Ed. 2009,
48, 1296.
(9) Thibaudeau, S.; Gouverneur, V. Org. Lett. 2003, 5, 4891.
(10) Suginome, M.; Ito, Y. Chem. ReV. 2000, 100, 3221.
(11) Horn, K. A. Chem. ReV. 1995, 95, 1317.
(12) Tsuji, Y.; Funato, M.; Ozawa, M.; Ogiyama, H.; Kajita, S.; Kawamura, T.
J. Org. Chem. 1996, 61, 5779.
Figure 2. Proposed catalytic cycle for silylation of allylic alcohols.
such a species is also promoted by electron donation from Pd(0)
to the π*-MO of the double bond, which increases the nucleophi-
licity of the oxygen atom. In complex 10 the C-O bond is
sufficiently weakened for an oxidative addition of Pd(0) to generate
allyl-palladium complex 11.
(13) Macsa´ri, I.; Hupe, E.; Szabo´, K. J. J. Org. Chem. 1999, 64, 9547.
(14) Moser, R.; Nishikata, T.; Lipshutz, B. H. Org. Lett. 2010, 12, 28.
(15) Suginome, M.; Ohmura, T.; Miyake, Y.; Mitani, S.; Ito, Y.; Murakami,
M. J. Am. Chem. Soc. 2003, 125, 11174.
(16) Wu, J. Y.; Stanzl, B. N.; Ritter, T. J. Am. Chem. Soc. 2010, 132, 13214.
(17) Hayashi, S.; Hirano, K.; Yorimitsu, H.; Oshima, K. J. Am. Chem. Soc.
2007, 129, 12650.
In allyl-palladium complex 11 the counterion is possibly a
hypervalent silylsilane arising from the C-O bond cleavage of 10.
This silane may easily transmetalate with palladium to give complex
13. The transmetalation delivers the silyl group to palladium, which
subsequently may undergo reductive elimination without any further
activation.13 Our previous studies13 have shown that transmetalation
of disilanes to allyl-palladium complexes and the subsequent
reductive elimination proceed very fast, with the high regioselec-
tivity affording the linear allylic isomer. In the reductive elimination
mechanism, the silyl group attacks the allyl moiety from the
direction of palladium via a syn mechanism. Supposing that the
allyl-palladium complex (11) forms mainly by the usual anti
stereochemistry and the nucleophilic attack proceeds by a syn
mechanism,41,42 a predominant inversion found for silylation of
1k to 4k can be easily explained. We do not exclude the possibility
that the mechanism of the presented borylation reaction (Figure 1)
is the same. Probably, the OH-activation29 and/or transmetalation
is more efficient for the borylation process than for the silylation
reaction, which may explain a faster formation of allylboronate 6a
than that of allylsilane 4a. Exploration of the exact mechanism
requires further experimental and modeling studies.
In summary, we have presented a new, efficient procedure for
the inexpensive and environmentally benign selective synthesis of
allylsilanes from allylic alcohols. The synthetic scope involves a
wide range of substituted allylic alcohols with internal and external
double bonds and even cyclic substrates. The applied reaction
conditions provide an easy access to allylic boronates as well. By
this study we have created a common platform for the synthesis of
functionalized allylsilanes and allylboronates from allylic alcohols
and the corresponding dimetallic reagents under identical reaction
conditions. Thus, here we provide a simple and efficient route to a
broad variety of building blocks for the synthesis of complex
organic molecules including natural products.2,43-45
(18) Simon, M. O.; Martinez, R.; Geneˆt, J. P.; Darses, S. AdV. Synth. Catal.
2009, 351, 153.
(19) Shintani, R.; Ichikawa, Y.; Hayashi, T.; Chen, J.; Nakao, Y.; Hiyama, T.
Org. Lett. 2007, 9, 4643.
(20) Taylor, R. E.; Engelhardt, F. C.; Schmitt, M. J.; Yuan, H. J. Am. Chem.
Soc. 2001, 123, 2964.
(21) Terao, J.; Watabe, H.; Watanabe, H.; Kambe, N. AdV. Synth. Catal. 2004,
346, 1674.
(22) Bourque, L. E.; Cleary, P. A.; Woerpel, K. A. J. Am. Chem. Soc. 2007,
129, 12602.
(23) Tsuji, Y.; Kajita, S.; Isobe, S.; Funato, M. J. Org. Chem. 1993, 58, 3607.
(24) Oestreich, M.; Auer, G. AdV. Synth. Catal. 2005, 347, 637.
(25) Kabalka, G. W.; Venkataiah, B.; Dong, G. Organometallics 2005, 24, 762.
(26) Beletskaya, I.; Moberg, C. Chem. ReV. 2006, 106, 2320.
(27) Muzart, J. Tetrahedron 2005, 61, 4179.
(28) Tamaru, Y. Eur. J. Org. Chem. 2005, 2647.
(29) Selander, N.; Kipke, A.; Sebelius, S.; Szabo´, K. J. J. Am. Chem. Soc. 2007,
129, 13723.
(30) Sebelius, S.; Olsson, V. J.; Wallner, O. A.; Szabo´, K. J. J. Am. Chem. Soc.
2006, 128, 8150.
(31) Olsson, V. J.; Sebelius, S.; Selander, N.; Szabo´, K. J. J. Am. Chem. Soc.
2006, 128, 4588.
(32) Selander, N.; Szabo´, K. J. J. Org. Chem. 2009, 74, 5695.
(33) Selander, N.; Szabo´, K. J. Dalton Trans. 2009, 6267.
(34) Dutheuil, G.; Selander, N.; Szabo´, K. J.; Aggarwal, V. K. Synthesis 2008,
2293.
(35) Sebelius, S.; Wallner, O. A.; Szabo´, K. J. Org. Lett. 2003, 5, 3065.
(36) Sebelius, S.; Szabo´, K. J. Eur. J. Org. Chem. 2005, 2539.
(37) Selander, N.; Sebelius, S.; Estay, C.; Szabo´, K. J. Eur. J. Org. Chem. 2006,
4085.
(38) Defieber, C.; Ariger, M. A.; Moriel, P.; Carreira, E. M. Angew. Chem.,
Int. Ed. 2007, 46, 3139.
(39) Kira, M.; Sato, K.; Kabuto, C.; Sakurai, H. J. Am. Chem. Soc. 1989, 111,
3747.
(40) Kano, N.; Nakagawa, N.; Shinozaki, Y.; Kawashima, T.; Sato, Y.; Naruse,
Y.; Inagaki, S. Organometallics 2005, 24, 2823.
(41) Hartwig, J. Organotransition Metal Chemistry; University Science Books:
Sausalito, CA, 2010.
(42) Kurosawa, H.; Ogoshi, S.; Kawasaki, Y.; Murai, S.; Miyoshi, M.; Ikeda,
I. J. J. Am. Chem. Soc. 1990, 112, 2813.
(43) Gawronski, J.; Wascinska, N.; Gajewy, J. Chem. ReV. 2008, 108, 5227.
(44) Kennedy, J. W. J.; Hall, D. G. Angew. Chem., Int. Ed. 2003, 42, 4732.
(45) Denmark, S. E.; Fu, J. Chem. ReV. 2003, 103, 2763.
JA1096732
9
J. AM. CHEM. SOC. VOL. 133, NO. 3, 2011 411