Stereoselective silylcupration of conjugated alkynes in water at room temperature

Article abstract of DOI:10.1002/anie.201311035

Micellar catalysis enables copper-catalyzed silylcupration of a variety of electron-deficient alkynes, thereby providing access to isomerically pure E- or Z-β-silyl-substituted carbonyl derivatives. These reactions take place in minutes, afford high yields and stereoselectivity, and are especially tolerant of functional groups present in the substrates. The aqueous reaction medium has been successfully recycled several times, and a substrate/catalyst ratio of 10,000:1 has been documented for this methodology. Fast, cheap, and green: Micellar catalysis enables the selective construction of a variety β-silyl-substituted carbonyl derivatives under mild aqueous conditions. The reaction is catalyzed by low levels of Cu^I, is compatible with numerous electron-withdrawing groups, affords high yields, and provides opportunities for scale-up and recycling of the reaction medium. The environmental impact, as measured by E Factors, is very low.

Full text of DOI:10.1002/anie.201311035

Angewandte
Chemie
DOI: 10.1002/anie.201311035
Green Chemistry
Stereoselective Silylcupration of Conjugated Alkynes in Water at
Room Temperature**
Roscoe T. H. Linstadt, Carl A. Peterson, Daniel J. Lippincott, Carina I. Jette, and
Bruce H. Lipshutz*
Abstract: Micellar catalysis enables copper-catalyzed silylcup-
ration of a variety of electron-deficient alkynes, thereby
providing access to isomerically pure E- or Z-b-silyl-substi-
tuted carbonyl derivatives. These reactions take place in
minutes, afford high yields and stereoselectivity, and are
especially tolerant of functional groups present in the sub-
strates. The aqueous reaction medium has been successfully
recycled several times, and a substrate/catalyst ratio of 10,000:1
has been documented for this methodology.
disclosed a method for the regioselective hydrosilylation of
internal alkynes using platinum catalysts.[20] Most recently,
and most closely analogous to the present work, Molander
described a copper-catalyzed introduction of silicon into
conjugated alkynes to give E-b-silyl-substituted carbonyl
derivatives. This system, however, gave variable stereoselec-
tivity and modest yields, and required the use of 1,3-dimethyl-
3,4,5,6-tetrahydro-2(1h)-pyrimidinone (DMPU) at relatively
high temperatures to activate a disilane to achieve conver-
sion.[21] Considering that copper-based reagents traditionally
undergo cis-addition to alkynes, and considering the ease with
which PhMe2SiBpin participates in transmetallation, we
sought to develop a complementary method to that of
Trost,[18] which would selectively provide entry to E-b-silyl-
substituted carbonyl derivatives under mild and, in particular,
greener conditions.
V
inylsilanes enjoy a rich history as valued intermediates in
organic synthesis.[1,2] Well established representative trans-
formations of this functional group include Hiyama–Den-
mark couplings,[3] iodode silylation,[4] and Brook rearrange-
ment/anion relay chemistry.[5] The use of vinylsilanes as
tethers or halogen precursors has been pivotal in several total
syntheses, in which they allow late stage functionalization.[6–8]
Traditionally, these substrates have been prepared by silyl-
cupration; for example, by using Flemingꢀs cuprate,[9] but such
processes often involve stoichiometric metals, and while these
may be of low cost, they carry prohibitive waste disposal
issues that usually preclude their use on a larger scale.[10]
Alternatively, Suginomeꢀs silylborane, PhMe2SiBpin, offers
the possibility of performing nucleophilic organosilicon
chemistry in water based on transmetallation from boron to
the appropriate transition metal.[11,12] Copper-catalyzed reac-
tions involving this reagent actually require the presence of
water or protic solvent for optimum efficiency.[13,14] Many
useful reactions have already been developed based on
silylboranes, as described in a comprehensive review by
Oestreich.[15] Traditional approaches to specifically E- or Z-b-
silylenoates rely on condensation of acylsilanes with yno-
lates[16] or phosphonate esters.[17] However, these approaches
suffer from a lack of efficiency and require multiple steps. A
more atom- and step-economical approach that has gained
popularity in recent years is the catalytic silylation of an
activated alkyne. Trost had shown previously that ruthenium
catalysts perform hydrosilylations to give Z-b-vinylsilanes.[18]
Palladium catalysts were recently found to give highly
selective a-silylation of conjugated alkynes,[19] and Ferreira
The initial reaction of ynoate 1 was performed with
PhMe2SiBpin as the stoichiometric source of silicon. Also
present were catalytic amounts of the air stable copper(I)
source CuF·(PPh3)3·2MeOH and bisphosphine ligand BDP,[22]
while an aqueous solution of the commercially available
surfactant TPGS-750-M (Figure 1) served as the reaction
Figure 1. Structure of TPGS-750-M and the BDP ligand.
medium. Under these room-temperature conditions, the E-b-
silylenoate (11) was generated as a single isomer (Table 1,
entry 1). Control reactions revealed that a copper(I) source is
required for the reaction, since no reaction took place in its
absence, even when this CuI salt was replaced with Cu-
(OAc)2·H2O (entries 2,5).[23] Likewise, attempts to use either
CuI or CuBr led to no conversion under otherwise identical
conditions (entries 3, 4).
Surprisingly, the use of CuI acetate was observed to give
full conversion after one hour (entry 6), thus suggesting that
the reaction was faster than originally anticipated and that
lower catalyst loadings could be employed. Performing the
reaction with ligand-free CuOAc led to limited conversion
(entry 7), confirming the benefits of a ligand on copper.
Reduced catalyst loading with the less expensive ligands
TMEDA and Ph3P gave roughly similar results (entries 8,9),
with the latter giving complete reaction in five minutes.
Interestingly, this model reaction could even be run “on
[*] R. T. H. Linstadt, C. A. Peterson, D. J. Lippincott, C. I. Jette,
Prof. B. H. Lipshutz
Department of Chemistry & Biochemistry, University of California
Santa Barbara, CA 93106 (USA)
E-mail: lipshutz@chem.ucsb.edu
[**] Financial support provided by the NSF is gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201311035.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Optimization of the reaction conditions.
Table 2: Substrate Scope: Acetylenic Esters.
Entry
Cu Souce
(mol%)
Ligand
(mol%)
Conc.
[m]
t
[h]
Conv.
[%]
1
2
3
4
5
6
7
8
9
10
CuF (3)[c]
none
CuI (3)
CuBr (3)
Cu(OAc)2·H2O (3)
CuIOAc (3)
CuIOAc (3)
CuIOAc (3)
CuIOAc (3)
CuIOAc (3)
BDP (3)
BDP (3)
BDP (3)
BDP (3)
BDP (3)
BDP (3)
none
TMEDA (3)
PPh3 (1)
PPh3 (1)
0.50
0.50
0.50
0.50
0.50
0.50
0.50
0.75
0.75
neat[d]
9
9
9
9
9
1
1
6
82[a]
<2[b]
0[a]
0[a]
0[a]
100[b]
44[b]
98[b]
100[b]
100[b]
5 min
5 min
Unless otherwise noted, all reactions were run under an inert
atmosphere in a 2 wt.% aqueous solution of TPGS-750-M with
1.25 equiv of PhMe2Si-Bpin. [a] Conversion monitored by GC–MS.
[b] Conversion monitored by NMR on the crude material. [c] Using
CuF(PPh3)3·2MeOH. [d] Using 5 equiv H2O.
water”, where the reactants, silylboronate, and copper
catalyst were mixed together neat with only five equivalents
of water (entry 10). Nevertheless, such “on water” reactions
are limited to mainly liquid substrates, while the use of
a surfactant offers flexibility by accommodating both solid
and liquid materials. Moreover, with greater volumes present,
opportunities exist for in-flask recycling of the aqueous
reaction medium.
With optimized conditions in hand (Scheme 1), the scope
of the reaction was next investigated with a variety of
acetylenic esters (Table 2). Several simple alkyl esters, such as
Scheme 1. Optimized conditions for reactions of acetylenic esters.
[a] With 5 mol% catalyst, 2 equiv PhMe2SiBpin, overnight, at room
temperature.
n-butyl (1), tert-butyl (2), and methyl (3) esters showed no
variation in terms of yield or stereochemical outcome. In the
case of extended conjugation such as that found in ynoate 4,
only the 1,4-addition product 14 was observed. Terminal
alkynes 5a and 5b afforded products 15a and 15b, respec-
tively, with complete E selectivity. A product enoate bearing
an alkyl chloride (16) could be obtained, albeit in modest
yield; increasing the amount of catalyst to 5% and the
silylboronate to two equivalents yielded the desired product
in 82% yield. The presence of an additional terminal alkyne
within ynoate 7 led to product 17 in 75% yield, with the
remaining mass balance being terminal alkyne addition
products and recovered starting material. To assess the effects
of an internal chelating residue on the stereochemical out-
come, a substrate with a 2-pyridyl group (8) was used and it
gave acceptable 12:1 E selectivity (entry 8). On the other
hand, a protected (9) or free (10) alcohol could be employed
with stereochemical impunity, although in the latter case,
spontaneous cyclization to the corresponding g-lactone (20)
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Table 3: Scope of alkynes bearing electron-withdrawing groups.
was observed, thus indicating that the reaction proceeded
with E selectivity.
Other electron-withdrawing groups on the alkyne were
then studied to evaluate the scope of the silylation (Table 3).
The acetylenic oxazolidinone 21 could be utilized to produce
33; the product was obtained predominantly as the E isomer
in 70% yield, along with the Z isomer (12%), each easily
separable by flash chromatography. The appearance of the
second isomer in this case suggests stabilization of an O-
chelated copper allenoate intermediate, prior to protio-
quenching (see below). Silylation of acetylenic acid 22 was
uneventful and gave at enoate 34 in 84% yield as a single
isomer. b-Silylation of acetylenic Weinreb amide 25 to give
product 37 represents a desirable transformation because it
offers routes to potential aldehyde and ketone derivatives. A
tertiary (26) or primary (28) amide could be used without
incident to give adducts 38 and 40, respectively. Alkynyl
peptides 23 and 24 led successfully to products 35 and 36, both
formed exclusively as E isomers. These conjugate additions
À
are thus tolerant of acidic N H functionality, and proceed
without erosion of the existing central chirality. The advant-
age of the surfactant becomes especially apparent in the case
of alkynyl peptide 23 and amide 28, where the reactants are
crystalline solids and cannot be used “on water.” Both
acetylenic sulfone 27 and nitrile 29 reacted smoothly in
> 95% yield, a notable result because the analogous vinylic
derivatives were previously reported to be sluggish towards
copper catalyzed silylation.[13]
Ketones were also investigated by using 30 as a model
substrate. The use of optimized conditions A (Table 3) led to
a disappointing 4:1 selectivity. However, to our surprise, 2D-
NOESY revealed that the Z-isomer was the major product.
With optimization, a combination of tri(p-fluorophenyl)phos-
phine-complexed copper(I) acetate, and performing the
reaction on water (conditions C), this value could be elevated
to give 42 as a 1:17 E/Z mixture.[24] Surprisingly, the analogous
isobutyl ketone led to lower selectivity (1:3 E/Z; 44) under
identical conditions. When the reaction was run in aqueous
surfactant with BDP as the ligand (conditions D) in an ice
bath, an acceptable 1:8 selectivity was obtained.[25] In contrast
to results from b-substituted ynones 30 and 32, terminal ynone
31 afforded product 43 with complete E selectivity, albeit in
modest (61%) yield.
To probe the mechanism of addition, the reaction was
conducted in aqueous surfactant solution derived from
TPGS-750-M dissolved in D2O. The selective deuterium
incorporation observed at the a-position implies that an a-
vinylcopper(I) species is formed and leads to eventual protio-
quenching[26] (see the Supporting Information). The high E
selectivity observed for esters likely arises from an unfavor-
able energetic barrier to enolization to the corresponding
isomeric O–Cu allenoate.[27,28]
While the base-metal status of copper, and hence its cost,
may not be a major factor governing its use, from an
environmental perspective, it remains desirable to reduce
the metal content in any catalytic process to the levels
dictated by necessity. The short reaction times suggested that
the level of copper required for this catalytic process could be
lowered considerably. This proved to be the case, since b-
Conditions A: PhMe2SiBpin (1.25 equiv)„ TPGS-750-M (2 wt.%), PPh3/
CuOAc (1 mol%), [0.75m substrate]. Conditions B: PhMe2SiBpin
(1.25 equiv), neat, H2O (5 equiv), PPh3/CuOAc (1 mol%). Conditions C:
PhMe2SiBpin (1.50 equiv), on water, [0.3m substrate], (4-F-C6H4)3P/
CuOAc (2 mol%). Conditions D: PhMe2SiBpin (1.00 equiv), ynoate
(2.00 equiv), [0.75m substrate], TPGS-750-M (2 wt.%), (4-F-C6H4)3P/
CuOAc (2 mol%). Conditions E: 1.25 equiv PhMe2SiBpin, 1 mol% BDP/
CuOAc, [0.75m substrate], TPGS-750-M (2 wt.%), ice bath temperature.
[a] Unless stated otherwise, all reactions were performed at ambient
temperature. [b] Using 2 equiv PhMe2SiBpin.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
silylation of 1 could be achieved with only 0.01% catalyst-to-
substrate loading (i.e., S/C = 10,000:1), in which case 16 h
were required for complete conversion (Scheme 2).
values are only slightly elevated to 8.5 and 2.4 when the water
used in the reaction medium is included.[30]
The product b-silylenoates can serve as precursors to
a variety of useful secondary products. For example, we have
previously demonstrated that nonracemically ligated copper
hydride can reduce E or Z b-silylenoates to afford chiral b-
silanes with excellent enantioselectivity.[31] Alternatively,
these substrates can be used to construct stereochemically
pure vinyl iodides. DIBAL-H reduction of 12 afforded the
corresponding b-silyl allylic alcohol (82%; Scheme 5). TBS-
protection and then iodination according to Vilarrasaꢀs
modification[32] of Zakarianꢀs conditions[4] gave E-vinyl
iodide (47) in 56% yield over four steps from ynoate 2 with
complete retention of stereochemistry.
Scheme 2. Reaction with 0.01 mol% catalyst loading.
Recycling of the aqueous reaction medium containing the
surfactant allows for even further reduction of aqueous and
metal waste, in addition to minimizing the use of organic
solvents. A simple in-flask extraction with a minimum amount
of organic solvent leads to isolation of the product. The
introduction of fresh substrate and silylborane to the recycled
aqueous medium, without additional surfactant or copper
catalyst, allows the reaction to be performed sequentially for
at least six cycles (Scheme 3). For each of these runs, the
product was isolated in excellent yield without adverse effects
regarding the reaction time or the purity of the product.
Scheme 5. Production of vinyl iodide 47. DIBAL-H=diisobutylalumi-
num hydride, TBS=tert-butyldimethylsilyl, NIS=N-iodosuccinimide,
HFIP=1,1,1,3,3,3-hexafluoro-2-propanol.
In summary, a new method has been developed that
allows especially facile copper-catalyzed silylcupration of
electron-deficient alkynes in water at room temperature.
Isomerically enriched b-silyl-substituted carbonyl derivatives
are readily formed quickly and in high yields. The process is
compatible with an array of electron-withdrawing groups and
consistently provides the anticipated geometrical isomer.
Considering the value of vinylsilanes in organic synthesis,
together with the overall reaction efficiency, selectivity,
options for scale-up, recyclability, and the environmentally
benign conditions involved, this chemistry should find con-
siderable utility.
Scheme 3. In-flask recycling of the catalyst and reaction medium.
Although these reactions were routinely performed on
a 0.1 to 0.4 mmol scale, these reactions are amenable to scale-
up. Starting with 3.94 mmol of substrate 2, 1.184 grams (86%)
of product 12 could be formed in less than 15 min (Scheme 4),
with the remaining mass being recovered starting material.
E Factors have become a commonly used metric to
evaluate the environmental impact of a given reaction.[29]
They are commonly defined as the ratio of the mass of
waste to the mass of desired product formed, and vary
depending upon the nature of the industry involved. Histor-
ically, only organic solvents, and not water, are used in the
calculation because the inclusion of water can significantly
increase these already elevated values. Nevertheless, for the
representative silylation of 1 to enoate 11, run in aqueous
surfactant (conditions A) or neat (conditions B), the E
Factors are calculated to be 4.2 and 2.1 respectively. These
Received: December 19, 2013
Published online: && &&, &&&&
Keywords: copper · green chemistry · micellar catalysis ·
.
organosilicon compounds · silylboronates
[1] I. Fleming, J. Dunoguꢁs, R. Smithers in Organic Reactions,
Wiley, New York, 2004, pp. 57 – 575.
[2] D. S. W. Lim, E. A. Anderson, Synthesis 2012, 44, 983.
[3] S. E. Denmark, J. M. Kallemeyn, J. Am. Chem. Soc. 2006, 128,
15958.
[4] E. A. Ilardi, C. E. Stivala, A. Zakarian, Org. Lett. 2008, 10, 1727.
[5] A. Tsubouchi, M. Itoh, K. Onishi, T. Takeda, Synthesis 2004,
1504.
[6] I. Volchkov, D. Lee, J. Am. Chem. Soc. 2013, 135, 5324.
[7] K. C. Nicolaou, X. Jiang, P. J. Lindsay-Scott, A. Corbu, S.
Yamashiro, A. Bacconi, V. M. Fowler, Angew. Chem. 2011, 123,
1171; Angew. Chem. Int. Ed. 2011, 50, 1139.
[8] A. T. Herrmann, S. R. Martinez, A. Zakarian, Org. Lett. 2011,
13, 3636.
[9] Lithium Bis[dimethyl(phenyl)silyl]cuprate. e-EROS Encyclope-
dia of Reagents for Organic Synthesis. [Online]; Wiley, New
Scheme 4. Gram-scale silylation.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
York, Posted April 15, 2001. http://onlinelibrary.wiley.com/doi/
10.1002/047084289X.rl054/full (accessed Oct 11, 2013).
[10] B. H. Lipshutz, Acc. Chem. Res. 1997, 30, 277.
[11] M. Suginome, T. Matsuda, Y. Ito, Organometallics 2000, 19, 4647.
[12]
T. Ohmura, H. Taniguchi, M. Suginome, J. Am. Chem. Soc. 2006,
128, 13682.
[13] J. A. Calderone, W. L. Santos, Org. Lett. 2012, 14, 2090.
[14] P. Wang, X. L. Yeo, T. P. Loh, J. Am. Chem. Soc. 2011, 133, 1254.
[15] M. Oestreich, E. Hartmann, M. Mewald, Chem. Rev. 2013, 113,
402.
[16] M. Shindo, K. Matsumoto, S. Mori, K. Shishido, J. Am. Chem.
Soc. 2002, 124, 6840.
[17] T. Fujiwara, K. Sawabe, T. Takeda, Tetrahedron 1997, 53, 8349.
[18] B. M. Trost, Z. T. Ball, J. Am. Chem. Soc. 2004, 126, 13942.
[19] Y. Sumida, T. Kato, S. Yoshida, T. Hosoya, Org. Lett. 2012, 14,
1552.
[20] D. A. Rooke, E. M. Ferreira, Angew. Chem. 2012, 124, 3279;
Angew. Chem. Int. Ed. 2012, 51, 3225.
[21] L. Iannazzo, G. A. Molander, Eur. J. Org. Chem. 2012, 4923.
[22] BDP = 1,2-bis(diphenylphosphino)benzene, see: B. A. Baker,
[25] Ice bath temperature; no freezing of the solution was observed.
[26] Attempts to trap this postulated a-cuprio intermediate with
electrophiles other than deuterium (e.g., allylic carbonates) were
unsuccessful.
[27] The energies for these types of intermediates have been
computationally examined for the analogous copper catalyzed
ˇ
´
hydroboration of ynoates. See: B. H. Lipshutz, Z. V. Boskovic,
D. H. Aue, Angew. Chem. 2008, 120, 10337; Angew. Chem. Int.
Ed. 2008, 47, 10183.
[28] b-Selectivity was confirmed by the presence of vinylic singlets in
the 1H spectra for all compounds possessing allylic hydrogens. E-
Selectivity for the esters, while implicit in the formation of
lactone 20, was confirmed by J-coupling constants for products
15a and 15b, which possess a b-hydrogen. Additional 1H–1H
NOESY spectra for compounds 13 and 14 confirmed
E
selectivity for these esters, and then by analogy, for all other
esters. 1H–1H NOESY was also performed for products E-33, Z-
33, 34, 35, 38, 39, 40, 41, 42, 43, and 44 to confirm the reported
stereochemistry; see the Supporting Information for further
details.
[29] R. A. Sheldon, Green Chem. 2007, 9, 1261.
ˇ
´
Z. V. Boskovic, B. H. Lipshutz, Org. Lett. 2008, 10, 289.
[23] CuII sources could be used if first mixed with 2 equiv of
a phosphine ligand (relative to Cu), thus suggesting that the
phosphine reduces CuII to the catalytically active CuI. For details
see the Supporting Information.
[30] B. H. Lipshutz, N. A. Isley, J. C. Fennewald, E. D. Slack, Angew.
Chem. 2013, 125, 11156; Angew. Chem. Int. Ed. 2013, 52, 10952.
[31] B. H. Lipshutz, N. Tanaka, B. R. Taft, C.-T. Lee, Org. Lett. 2006,
8, 1963.
[32] M. Sidera, A. M. Costa, J. Vilarrasa, Org. Lett. 2011, 13, 4934.
[24] See the Supporting Information for details regarding optimiza-
tion of the reaction conditions for alkynyl ketones.
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Green Chemistry
R. T. H. Linstadt, C. A. Peterson,
D. J. Lippincott, C. I. Jette,
B. H. Lipshutz*
&&&& — &&&&
Stereoselective Silylcupration of
Conjugated Alkynes in Water at Room
Temperature
Fast, cheap, and green: Micellar catalysis
enables the selective construction of
a variety b-silyl-substituted carbonyl
derivatives under mild aqueous condi-
tions. The reaction is catalyzed by low
levels of CuI, is compatible with numer-
ous electron-withdrawing groups, affords
high yields, and provides opportunities
for scale-up and recycling of the reaction
medium. The environmental impact, as
measured by E Factors, is very low.
6
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
These are not the final page numbers!