Highly β-(E)-selective hydrosilylation of terminal and internal alkynes catalyzed by a (IPr)Pt(diene) complex

Article abstract of DOI:10.1021/jo800411e

(Chemical Equation Presented) The regioselective hydrosilylation of terminal and internal alkynes catalyzed by the novel (IPr)Pt(AE) (7) (IPr = bis(2,6-diisopropylphenyl)imidazo-2-ylidene, AE = allyl ether) complex is presented. The (IPr)Pt(AE) catalyst displays enhanced activity and regioselectivity for the hydrosilylation of terminal and internal alkynes with low catalyst loading (0.1 to 0.05 mol %) when compared to the parent (IPr)Pt(DVDS) complex (6) (DVDS = divinyltetramethyldisiloxane). The reaction leads to exquisite regioselectivity in favor of the cis-addition product on the less hindered terminus of terminal and internal alkynes. The solvent effects were examined for the difficult hydrosilylation of benzylpropargyl ether. In light of the observed product distribution and kinetic data, a mechanistic scheme is proposed involving two competing catalytic cycles. One cycle leads to high regioselectivities while the other, having lost the stereodirecting IPr carbene ligand, displays low regiocontrol and activities. The importance of this secondary catalytic cycle is either caused by the strong coordinating ability of the alkyne or by the low reactivity of the silane or both.

Full text of DOI:10.1021/jo800411e

Highly ꢀ-(E)-Selective Hydrosilylation of Terminal and Internal
Alkynes Catalyzed by a (IPr)Pt(diene) Complex
Guillaume Berthon-Gelloz, Jean-Marc Schumers, Guillaume De Bo, and Istva´n E. Marko´*
UniVersite´ catholique de LouVain, De´partement de Chimie, Place Louis Pasteur,
1348 LouVain-la-NeuVe, Belgium
istVan.marko@uclouVain.be
ReceiVed February 20, 2008
The regioselective hydrosilylation of terminal and internal alkynes catalyzed by the novel (IPr)Pt(AE)
(7) (IPr ) bis(2,6-diisopropylphenyl)imidazo-2-ylidene, AE ) allyl ether) complex is presented. The
(IPr)Pt(AE) catalyst displays enhanced activity and regioselectivity for the hydrosilylation of terminal
and internal alkynes with low catalyst loading (0.1 to 0.05 mol %) when compared to the parent
(IPr)Pt(DVDS) complex (6) (DVDS ) divinyltetramethyldisiloxane). The reaction leads to exquisite
regioselectivity in favor of the cis-addition product on the less hindered terminus of terminal and internal
alkynes. The solvent effects were examined for the difficult hydrosilylation of benzylpropargyl ether. In
light of the observed product distribution and kinetic data, a mechanistic scheme is proposed involving
two competing catalytic cycles. One cycle leads to high regioselectivities while the other, having lost the
stereodirecting IPr carbene ligand, displays low regiocontrol and activities. The importance of this
secondary catalytic cycle is either caused by the strong coordinating ability of the alkyne or by the low
reactivity of the silane or both.
Introduction
coupling reactions.3–5 Vinylsilanes can also act as masked
carbonyl groups, being ultimately revealed by a Tamao-Fleming
oxidation.6–8 For both of these processes to readily occur, at
least one heteroatom, such as F, Cl, or O, is required on the
silicon atom.
The most straightforward and atom-economical manner to
assemble vinylsilanes is by the regioselective hydrosilylation
of alkynes. Trost et al. have reported a highly regioselective,
ruthenium-based system, which promotes trans addition of
silanes to terminal and internal alkynes to form the correspond-
ing R isomers 1 and 2, respectively.9,10 The trans addition
product (3, ꢀ-(Z)) can be accessed by rhodium11,12 and ruthe-
Over the past decades, vinylsilanes have emerged as versatile
building blocks in organic synthesis.1,2 Their low cost, high
stability, and innocuous byproduct have made them attractive
surrogates for their tin- and boron-derived counterparts. A
significant advantage of vinylsilanes over most other vinylmetal
species is their ability to be carried through many synthetic steps
without decomposition. One of their most promising applications
is their use as nucleophilic partners in palladium-catalyzed cross-
* To whom correspondence should be addressed. Phone: 32/10478773. Fax:
+32/10472788.
(1) Ojima, I.; Li, Z. Y.; Zhu, J. W. In Chemistry of Organic Silicon
Compounds; Rappoport, Z., Apeloig, Y. Eds.; John Wiley & Sons Ltd:
Chichester, England, 1998; pp 1687-1792.
(6) Fleming, I.; Henning, R.; Plaut, H. Chem. Commun. 1984, 29, 30.
(7) Tamao, K.; Maeda, K. Tetrahedron Lett. 1986, 27, 65–68.
(8) Jones, G. R.; Landais, Y. Tetrahedron 1996, 52, 7599–7662.
(9) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2001, 123, 12726–12727.
(10) Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2005, 127, 17644.
(11) Ojima, I.; Clos, N.; Donovan, R. J.; Ingallina, P. Organometallics 1990,
9, 3127.
(2) Langkopf, E.; Schinzer, D. Chem. ReV. 1995, 95, 1375–1408.
(3) Hiyama, T. In Metal-Catalyzed Cross-Coupling Reactions; Diederich,
F., Stang, P. J. Eds.; Wiley-VCH, Inc: Weinheim, Germany, 1998; pp 421-
453.
(4) Hiyama, T.; Shirakawa, E. Top. Curr. Chem. 2002, 219, 61–85.
(5) Denmark, S. E.; Sweis, R. F. In Metal-Catalyzed Cross-Coupling
Reactions, 2nd ed.; de Meijere, A., Diederich, F. Eds.; Wiley-VCH: Weinheim,
Germany, 2004, p 163-216.
(12) Doyle, M. P.; High, K. G.; Nesloney, C. L.; Clayton, J. W., Jr.; Lin, J.
Organometallics 1991, 10, 1225.
4190 J. Org. Chem. 2008, 73, 4190–4197
10.1021/jo800411e CCC: $40.75 2008 American Chemical Society
Published on Web 05/14/2008

Highly ꢀ-(E)-SelectiVe Hydrosilylation of Alkynes
nium13 catalyzed hydrosilylations, even though the silanes
employed remained limited to Et3SiH and Ph3SiH, except for
one notable exception reported by Hiyama and Mori.14
SCHEME 1. Advantages and Disadvantages of MDHM as
an Industrial Silan
SCHEME 2. Synthetic Relevance of the -SiMe(OSiMe3)2
(DM2) Group in a Cross-Coupling Reaction
On the other hand, the platinum-catalyzed hydrosilylation of
alkynes has been widely investigated and shown to proceed by
the cis addition of the hydrosilane across the alkyne, forming
exclusively adducts 1, 4, and 5.15,16 High selectivities in favor
of the ꢀ-(E) isomer have been reported with use of platinum
complexes bearing bulky phosphane ligands.17–25 However,
closer examination reveals that only a limited range of silanes,
such as chloro-, trialkyl-, and triarylsilanes, truly offer high ꢀ-(E)
regioselectivities. In contrast, alkoxysilanes are usually poor
substrates for the Pt-catalyzed hydrosilylation of alkynes.26
Chlorosilanes give high regioselectivites; however, they are toxic
and delicate to handle, and the chlorosilyl function cannot be
carried through several synthetic steps without prior modifica-
tion. Finally, trialkyl- and triarylvinysilanes provide limited
synthetic utility.1 To overcome these issues, masked silanols
have been introduced, such as PhMe2SiH,27 2-thienyldimeth-
ylsilane,28 2-pyridyldimethylsilane,29 and BnMe2SiH,30 but these
silanes are relatively onerous. A silane, which has remained
underutilized in organic chemistry, is bis(trimethylsilyloxy)m-
ethylsilane (MDHM), an industrial compound produced on the
ton-scale.31 MDHM combines the benefits of low cost and high
hydrolytic stability (due to its siloxane linkage, Si-O-Si-O-Si).
Moreover, MDHM displays a low reactivity akin to that of
dialkoxysilanes and is also less bulky than its structure would
imply (e.g., smaller than Et3SiH) (Scheme 1).
philic partner in Hiyama-type cross-coupling under standard
conditions (Scheme 2).
It thus transpires that off-the-shelf catalytic systems, which
can achieve high yields of the ꢀ-(E) isomer with synthetically
useful silanes, are still needed. Platinum(0) N-heterocyclic
carbene (NHC) complexes are significantly more stable than
their corresponding phosphine analogues and can be stored in
the solid form and in solution, under ambient conditions, without
significant decomposition. In our previous report on the hy-
drosilylation of alkynes mediated by N-heterocyclic carbene
platinum(0) complexes, we have identified the (IPr)Pt(DVDS)
(IPr ) 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene; DVDS
) divinyltetramethyldisiloxane) complex (6) as the most active
catalyst of the series investigated.32 Thus, the hydrosilylation
of 1-octyne by MDHM, in the presence of (IPr)Pt(DVDS) (5 ×
10-3 mol %; 80 °C; 6 h), yielded the ꢀ-(E) isomer in a 10.6:1
ratio with the R isomer, in greater than 99% yield. Although
this result was promising, it was still lagging behind existing
methodologies based upon bulky phosphine ligands.
Herein, we wish to report a modification of our previously
described catalytic system for the hydrosilylation of alkynes
which displays highly enhanced activity, superior ꢀ-(E) selectiv-
ity, and broader substrate scope. We also address the little
developed regioselective cis-hydrosilylation of internal alkynes.
A study of the initiation period has enabled us to delineate key
mechanistic features and to pinpoint several deactivation
pathways occurring in our system.
Denmark et al. have demonstrated that related disiloxanes
give high yields in cross-coupling reactions20 and preliminary
results from our laboratory have shown that the high stability
of the Si-O-Si bonds does not preclude its use as a nucleo-
(13) Na, Y.; Chang, S. Org. Lett. 2000, 2, 1887–1889.
(14) Mori, A.; Takahisa, E.; Yamamura, Y.; Kato, T.; Mudalige, A. P.; Kajiro,
H.; Hirabayashi, K.; Nishihara, Y.; Hiyama, T. Organometallics 2004, 23, 1755–
1765.
Results and Discussion
(15) Speier, J. L. AdV. Organomet. Chem. 1979, 17, 407–447.
(16) Lewis, L. N.; Sy, K. G.; Bryant, G. L., Jr.; Donahue, P. E. Organome-
tallics 1991, 10, 3750–3759.
Optimization of Hydrosilylation Conditions with Catalyst
7. The (IPr)Pt(DVDS) complex (6) suffered from a lack of
catalytic activity (although it was the most active among
(NHC)Pt(DVDS) complexes investigated) and therefore long
reaction times were required. Relatively high temperatures (80
°C) were also necessary to effect complete conversions. Our
previous studies have revealed that the rate determining step in
the catalyst activation is the initial decoordination of the diene
ligand (Scheme 3).33 This is due to the presence of the DVDS
ligand which, being particularly strongly bound to the plati-
num(0) center, is difficult to displace to free the active [(IPr)Pt]
species (Scheme 3, I).32
(17) Green, M.; Spencer, J. L.; Stone, G. A. F.; Tsipis, C. A. J. Chem. Soc.,
Dalton Trans. 1977, 1525, 1529.
(18) Murphy, P. J.; Spencer, J. L.; Procter, G. Tetrahedron Lett. 1990, 31,
1051–1054.
(19) Takahashi, K.; Minami, T.; Ohara, Y.; Hiyama, T. Tetrahedron Lett.
1993, 34, 8263–8266.
(20) Denmark, S. E.; Wang, Z. Org. Lett. 2001, 3, 1073–1076.
(21) Itami, K.; Mitsudo, K.; Nishino, A.; Yoshida, J.-I. J. Org. Chem. 2002,
67, 2645–2652.
(22) Wu, W.; Li, C.-J. Chem. Commun. 2003, 1668, 1669.
(23) Aneetha, H.; Wu, W.; Verkade, J. G. Organometallics 2005, 24, 2590–
2596.
(24) Hamze, A.; Provot, O.; Brion, J. D.; Alami, M. Synthesis 2007, 2025,
2036.
(25) Hamze, A.; Provot, O.; Brion, J.-D.; Alami, M. Tetrahedron Lett. 2008,
49, 2429–2431.
It was therefore anticipated that the replacement of the DVDS
ligand by a more labile allyl ether (AE) moiety would increase
(26) Trost, B. M.; Ball, Z. T. Synthesis 2005, 6, 853–887.
(27) Anderson, J. C.; Munday, R. H. J. Org. Chem. 2004, 69, 8971–8974.
(28) Hosoi, K.; Nozaki, K.; Hiyama, T. Chem. Lett. 2002, 31, 138–139.
(29) Itami, K.; Nokami, T.; Yoshida, J.-I. J. Am. Chem. Soc. 2001, 123, 5600–
5601.
(32) De Bo, G.; Berthon-Gelloz, G.; Tinant, B.; Marko´, I. E. Organometallics
2006, 25, 1881–1890.
(30) Trost, B. M.; Machacek, M. R.; Ball, Z. T. Org. Lett. 2003, 5, 1895–
1898.
(33) Berthon-Gelloz, G.; Marko´, I. E. In N-Heterocyclic Carbenes in
Synthesis; Nolan, S. P. Ed.; Wiley-VCH: Weinheim,Germany, 2006; pp 119-
161.
(31) MDHM is produced for about 10 Eur/kg.
J. Org. Chem. Vol. 73, No. 11, 2008 4191

Berthon-Gelloz et al.
SCHEME 3. Proposed Initiation Mechanism for
(NHC)Pt(diene) Complexes
FIGURE 1. (NHC)Pt(diene) complexes 6 and 7.
SCHEME 4. Synthesis of (IPr)Pt(AE) (7)
characteristic yellow taint, signaling the formation of a catalyti-
cally active species. Hydrosilylations, in which this yellow color
did not appear, afforded low conversions and poor selectivities.
The optimization study revealed that the temperature and the
catalyst loading were crucial parameters. It was necessary to
perform the reaction above 40 °C to maintain a good catalytic
activity. This is especially true at lower catalyst loadings (0.05
mol %). In several instances, the reaction stalled when effected
at 20 °C. Moreover, the hydrosilylation reaction proved to be
quite sensitive to catalyst loading and a sharp decrease in activity
and selectivity was observed when 0.01 mol % of 7 was
employed instead of 0.05 mol % (Table 1, entries 5 and 6,
respectively). This dramatic reduction in reaction rate and
stereocontrol is presumably due to a specific deactivation of
the catalyst (vide infra). In accord with previous observations
made by Stone et al.,17 the use of solventless conditions
strikingly increased the velocity of this process, without affecting
the yield or the regioselectivity of the addition (Table 1, entry
7). This is particularly useful when performing the reaction on
a large scale, even though care must be exercised in controlling
the ensuing exotherm (vide infra).
the rate of formation of the catalytically active [(IPr)Pt] fragment
(Figure 1). Similar effects of so-called “spectator” diene ligands
have become widely recognized in catalysis.34 Elegant studies
on the Pd2(dba)3 (dba ) trans,trans-dibenzylidene acetone)
precatalysts by Fairlamb et al. have shown that electronically
modified dba ligands significantly influence the overall catalytic
activity in cross-coupling reactions.35,36 The choice of the diallyl
ether fragment offers a good compromise between stability and
activity of the complex37 and was inspired by studies by
Po¨rschke et al., who first reported complex 8.38
The synthesis of the (IPr)Pt(AE) complex (7) was ac-
complished by treatment of a solution of Pt2(AE)3 with the IPr
carbene formed in situ (Scheme 4).38 A direct, high-yielding
synthesis of (IPr)Pt(AE) (7) and congeners from H2PtCl6 has
been recently reported.39 Complex 7 is an air-stable, off-white
solid that can be stored indefinitely under ambient conditions
and is stable as a toluene solution for several weeks.
With complex (IPr)Pt(AE) (7) in hand, we began to inves-
tigate its activity in the hydrosilylation reaction and chose, as a
model transformation, the addition of MDHM to 1-octyne (Table
1). To assess the base selectivity of this process, we employed
PtCl2(cod) (cod ) cis,cis-1,5-cyclooctadiene) as a reference
catalyst lacking any bulky regiodirecting ligands (Table 1, entry
1). Lewis et al. have shown that this platinum complex gives
the same regioselectivity as the more commonly used Karstetd’s
catalyst40 ([Pt(CH2dCHSiMe2)2O]).16 The optimum conditions
were obtained by carrying out the reaction at 60 °C, under
solventless conditions, in the presence of 0.1 mol % of 7. A
slight excess of silane (10%) was used (Table 1, entry 8) to
ensure reproducibility of the reaction. Under these conditions,
the reaction was essentially over within 5 min, affording the
ꢀ-(E) isomer in high yield and with exquisite selectivity. At
the onset of this transformation, the medium acquired a
Using the optimum hydrosilylation conditions, we compared
the activity of catalyst 7 with that of catalyst 6 bearing the
DVDS ligand (Table 1, entries 9 and 10). Catalyst 6 displayed
very little hydrosilylation activity at 25 °C and gave incomplete
conversion and moderate selectivity at 60 °C. This confirms
that the substitution of the DVDS ligands by the more labile
allyl ether is crucial for catalytic activity.
For example, in a separate experiment, 1-octyne was reacted
with MDHM on a 50 mmol scale, at 60 °C, under solventless
conditions. The addition of the catalyst 7 (5 × 10-2 mol %) caused
the reaction temperature to rise to 180 °C within 1 min. After 3
min, the hydrosilylation was complete. Interestingly, even at
this elevated temperature, the ꢀ-(E)/R ratio remained a remark-
able 25:1 (Scheme 5). The product was obtained in an isolated
yield of 95%, indicating the high stability of the catalytically
active [(IPr)Pt] fragment and the robustness of the present
catalytic system. Therefore, the regioselectivity of the addition
is mostly dictated by the steric bulk of the IPr ligand and by
the nature of the substrates; the temperature only plays a very
marginal role. Similar observations have been reported by
Yoshida et al. when they performed the hydrosilylation of
1-octyne by dimethyl(pyridyl)silane with (t-Bu3P)Pt(DVDS) (1
mol %) at 100 °C and still achieved high regioselectivity for
the ꢀ-(E)-vinylsilane.21 ObViously, care must be taken to aVoid
such strong exotherms! The exotherm can be controlled on a
large scale by performing the reaction at 30 °C with the reaction
vessel immersed in a water bath or by slow addition of the
alkyne to the mixture of catalyst and silane at room temperature.
Solvent Effect and Deactivation Pathway. The hydrosily-
lation of a more challenging substrate, benzylpropargyl ether
(34) Johnson, J. B.; Rovis, T. Angew. Chem., Int. Ed. 2008, 47, 840–871.
(35) Fairlamb, I. J. S.; Kapdi, A. R.; Lee, A. F. Org. Lett. 2004, 6, 4435–
4438.
(36) Fairlamb, I. J. S.; Kapdi, A. R.; Lee, A. F.; McGlacken, G. P.;
Weissburger, F.; H, M.; de Vries, A.; Schmieder-van de Vondervoort, L. Chem.
Eur. J. 2006, 12, 8750–8761.
(37) Related (NHC)Pt(norbornene)2 complex reported by Elsevier et al.
proved too unstable to be isolated and had to be stabilized by complexation
with maleic anhydride.
(38) Krause, J.; Cestaric, G.; Haack, K.-J.; Seevogel, K.; Storm, W.; Po¨rschke,
K.-R. J. Am. Chem. Soc. 1999, 121, 9807–9823.
(39) Berthon-Gelloz, G.; Schumers, J.-M.; Tinant, B.; Wouters, J.; Marko´,
I. E. Organometallics 2007, 26, 5731–5734.
(40) Karstedt, B. D. General Electric. U.S. Patent 3,715,334, 1970.
4192 J. Org. Chem. Vol. 73, No. 11, 2008

Highly ꢀ-(E)-SelectiVe Hydrosilylation of Alkynes
TABLE 1. Optimization of the Hydrosilylation of 1-Octyne by MDHM Catalyzed by 7a
entry
catalyst (mol %)
solvent
T (°C)
t (min)
ratio (ꢀ:R)b
yield (%)c
1
PtCl2(cod) (5 × 10-3
)
THF
o-Xy
THF
THF
THF
THF
60
80
25
25
40
60
25
60
60
25
60
10
3:1
11:1
100:1
50:1
200:1
11.5:1
100:1
200:1
100:1
75
90
90
80
84
38
95
97
98
<5
60
2d
3
6 (5 × 10-3
7 (0.1)
)
360
400
400
100
120
80
5
10
12 h
12 h
4
5
6
7 (0.05)
7 (0.05)
7 (0.01)
7 (0.1)
7 (0.1)
7 (0.05)
6 (0.1)
7e
8e
9e
10
11
-
f
f
-
f
-
f
-
f
6 (0.1)
-
10:1
a Reaction conditions: 1-octyne (3.0 mmol, 1 equiv), MDHM (3.0 mmol, 1 equiv), dodecane (internal standard), [Pt], solvent (1.0 M). b Ratio
e
determined by GC analysis of the crude reaction mixture. c GC yield of the ꢀ-(E) isomer. d Results reported in ref 30. 1.1 equiv of MDHM used.
f Solventless conditions. cod ) cis,cis-1,5-cyclooctadiene. -DM2: -SiMe(OSiMe3)2.
SCHEME 5. Robustness of the Catalytic System at Elevated
Temperatures
FIGURE 2. Factors making BnOP (10) a difficult substrate to
hydrosilylate.
SCHEME 6. Proposed Deactivation Pathway for
(IPr)Pt(AE) (7) in the Presence of Coordinating Alkynes
TABLE 2. Optimization of the Hydrosilylation Conditions of 10a
entry
7 (mol %)
solvent T (°C) t (h) ratio (ꢀ:R)b conv (%)c
1
2
3
4
5
6
7
8
9
PtCl2(cod) (0.1) THF
20
20
20
20
20
20
20
20
20
1
1.5:1
2.0:1
2.6:1
1.8:1
92
40
d
0.1
0.1
0.1
0.1
0.1
0.1
1.0
1.0
-
5 d
5 d
7 d
7 d
5 d
5 d
0.1
2
toluene
CH2Cl2
CH3CN
acetone
-PrOH
44 (47)
60 (73)
0
and acetone led to marginal improvements (Table 2, entries 3,
4, and 6, respectively), acetonitrile completely inhibited the
catalyst (Table 2, entry 5), presumably due to its strong
coordinating capability. However, performing the reaction in
isopropyl alcohol afforded adducts 11a and 11b with a 13.3:1
selectivity in favor of the ꢀ-(E) isomer 11a (Table 2, entry 7).
Surprisingly, the initially fast reaction rate quickly decreased,
leading to incomplete conversion even after 5 days. Concomi-
tantly, erosion of the selectivity was also observed and some
silylation of isopropyl alcohol was detected after several days.
The higher selectivity obtained in isopropyl alcohol, as compared
to other solvents, could be rationalized by invoking a hydrogen-
bonding interaction between the alcohol and the ether function
of BnOP. Such H-bonding would limit the availability of the
ether lone pair for coordination to the platinum center and hence
disfavor the formation of the unwanted R isomer 11b.
Eventually, the use of 1 mol % of the (IPr)Pt(AE) (7) catalyst
in THF gave the desired product 11a in a 98% isolated yield
and with a regioselectivity of 50:1 in favor of the ꢀ-(E) isomer
(Table 2, entry 9). Although these catalyst loadings are quite
high, similar loadings have been found necessary by Verkade
et al. for the hydrosilylation of BnOP by triethylsilane.23
Nevertheless, the hydrosilylation of BnOP gave us some useful
information about the catalytic system. High activities and
regioselectivities are achieved when the hydrosilylation occurs
2.3:1
13.3:1
15.7:1
50:1
40 (62)e
61 (88)f
94g
d
-
THF
98g
a Reaction conditions: BnOP (10) (1 equiv), PhMe2SiH (1 equiv),
dodecane (internal standard), solvent (1.0 M). b Ratio determined by GC
analysis of the crude reaction mixture. c GC conversion of the alkyne
(silane). d Solventless conditions. e Hydrosilylation of acetone occurred.
f Partial silylation of -PrOH occurred. g Isolated yield.
(BnOP, 10), by dimethylphenylsilane (Table 2) was next
investigated. BnOP combines a coordinating ether function with
a lower electronic density on the internal carbon of the alkyne,
both features greatly favoring the R isomer (Figure 2).41 Thus,
the hydrosilylation of BnOP by dimethyphenylsilane, catalyzed
by PtCl2(cod) affords the desired adduct with an extremely low
ꢀ-(E)/R ratio of 1.5:1 (Table 2, entry 1).
When the optimal conditions developed for the hydrosilylation
of 1-octyne where applied to BnOP, only poor conversion and
mediocre selectivities resulted, even after 5 days (Table 2, entry
2). Several solvents were screened to improve both regioselec-
tivity and catalytic activity. While toluene, dichloromethane,
(41) Tsipis, C. A. J. Organomet. Chem. 1980, 187, 427–446.
J. Org. Chem. Vol. 73, No. 11, 2008 4193

Berthon-Gelloz et al.
TABLE 4. Hydrosilylation of internal alkynes by complex 7a
TABLE 3. Hydrosilylation of Terminal Alkynes by Complex
(IPr)Pt(AE) 7a
a Reaction conditions: alkyne (1.0 equiv), silane (1.1 equiv),
(IPr)Pt(AE) (7) (0.1 mol %), solventless, 60 °C. b Time for complete
conversion by GC analysis. c Ratio of regioisomers determined by GC
analysis. d Isolated yield of the mixture of regioisomers. e Ratio
determined on the crude reaction mixture by 1H NMR. f 0.5 mol % of
catalyst was used.
faster than the catalyst deactivation. Hence, with 1 mol % of
(IPr)Pt(AE) (7), the reaction is extremely selective and efficient
while poor regioselectivities and conversions are observed with
0.1 mol %.
a Reaction conditions: alkyne (1.0 equiv), silane (1.1 equiv),
(IPr)Pt(AE) (7) (0.1 mol %), solventless, 60 °C. b Time for complete
conversion by GC analysis. c Ratio of regioisomers (major:minor)
determined by GC and 1H NMR analysis of the crude reaction mixture.
d Isolated yield of the mixture of regioisomers. e 0.5 mol % of catalyst
was used, THF (1.0 M). f Ratio determined on the crude reaction
The reduced regioselectivity observed with lower catalyst
loading hints to the fact that a concurrent hydrosilylation cycle
might be operating as a background reaction and that the active
species in this cycle could presumably be lacking the bulky IPr
ligand. It is thus possible that the strong coordination of BnOP
to platinum may lead to the displacement of the IPr ligand,
affording a Pt(η2-alkyne)2 complex that is a poor and unselective
hydrosilylation catalyst.42 To substantiate this hypothesis, (IP-
r)Pt(AE) was reacted with 5 equiv of BnOP (10) in toluene-d8
and the reaction was monitored by 1H NMR. After several hours
at 40 °C, the 1H NMR spectrum of the mixture displayed
complete decomposition of the complex. The signals belonging
to the coordinated allyl ether (3.95 and 2.44 ppm) had
completely disappeared and the region of the IPr methyl groups
displayed a multitude of peaks, indicating a complex mixture.
More significantly, signals corresponding to the protonated IPr
carbene appeared (9.9 ppm) probably due to the decoordination
of the IPr carbene from the metal, followed by its protonation
by residual water.
Substrate Scope and Limitation. The hydrosilylation of
alkynes with dimethylphenylsilane proceeded faster and with
comparable or higher selectivities than with MDHM (Table 3,
entries 1 and 3). Gratifyingly, propargyl alcohol was hydrosi-
lylated with remarkable efficiency with use of a catalyst loading
of 0.1 mol % (Table 3, entry 4). This stands in sharp contrast
to the difficulties encountered in the hydrosilylation of BnOP
(10). The transformation of pent-4-yn-ol into 16 (Table 3, entry
5) took place equally smoothly. Thus, it appears that hydrogen
mixture by H NMR. g 1 mol % of catalyst. -DM2 ) -SiMe(OSiMe3)2.
1
bonding in these systems plays a beneficial role in upholding
high regioselectivities, comfirming the previous observation that
higher ꢀ-(E)/R ratios were obtained in isopropyl alcohol. This
observation also holds true for a free amine function. Hence,
propargyl amine was hydrosilylated with a high regioselectivity,
albeit a catalyst loading of 0.5 mol % had to be used (Table 3,
entry 7). It has been previously reported that, in hydrosilylation
of propargylic alcohols with (t-Bu3)Pt(DVDS), the addition of
a catalytic amount of Na(0) appears to be necessary to achieve
high levels of regioselectivity. This additive is not necessary
with the present catalyst.43 In all these cases, no silylation of
the free alcohol or the free amine was observed. Trimethylsi-
lylacetylene, which is particularly reactive and gives poor
regioselectivities with PtCl2(cod) (ꢀ/R ratio of 1.5/1, due to the
presence of a silicon center), was also hydrosilylated with high
regioselectivity (Table 3, entry 8).
The high regioselectivities obtained for the hydrosilylation
of terminal alkynes prompted us to investigate the much more
challenging hydrosilylation of internal acetylenes catalyzed by
complex 7 (Table 4). Therefore, we examined the ability of our
system to discriminate between a simple methyl group and
another substituent on the triple bond. In the case of a linear
alkyl chain, the selectivity was modest with MDHM (Table 4,
(42) Boag, N. M.; Green, M.; Grove, D. M.; Howard, J. A. K.; Spencer,
J. L.; Stone, G. A. F. J. Chem. Soc., Dalton Trans. 1980, 2170.
(43) Beresis, R. T.; Solomon, J. S.; Yang, M. G.; Jain, N. F.; Panek, J. S.
Org. Synth., Collect. Vol. 2004, 10, 531.
4194 J. Org. Chem. Vol. 73, No. 11, 2008

Highly ꢀ-(E)-SelectiVe Hydrosilylation of Alkynes
FIGURE 3. Kinetic profile of the reaction of MDHM and 1-octyne
catalyzed by (IPr)Pt(AE) (7). Curve A: Conversion of 1-octyne. Curve
B: Formation of the ꢀ-(E) (9a) isomer. Reaction conditions: MDHM
(6.0 mmol, 1 equiv), 1-octyne (6.0 mmol, 1 equiv), dodecane (internal
standard), (IPr)Pt(AE) (7, 6 × 10-3 mmol, 0.1 mol %), THF (1.0 M),
20 °C. The evolution of the reaction was followed by GC analysis.
entry 1). It increased slightly when the more hindered silane,
Me2BnSiH, was employed (Table 4, entry 2) and nearly doubled
when PhMe2SiH was used as the hydrosilylating partner (Table
4, entry 3). This trend continued with the use of Ph2MeSiH
silane, which gave a 4.4:1 selectivity for the least hindered
isomer (Table 4, entry 4). This level of regioselectivity is on
par with the best cis hydrosilylating catalysts reported to date
([(Cp*)Y(CH3)(THF)]).44 Interestingly, the regioselectivity can
be further increased to 6.0:1 by employing Ph2ClSiH as a silane
(Table 4, entry 5), suggesting that chloro substituent on silicon
exerts a more powerful regiodirecting effect than a bulkier
methyl group. From these experiments, silanes can be ranked
in order of increasing steric demand against 2-nonyne: MDHM
< BnMe2SiH < PhMe2SiH < Ph2MeSiH < Ph2ClSiH. There-
fore, MDHM behaves as a rather small silane and thus obtaining
high selectivity with this silane is a real challenge.
As might have been expected, a tert-butyl substituent gave
useful selectivity (Table 4, entry 6). The hydrosilylation of
1-phenyl-2-propyne proceeded smoothly and efficiently (Table
4, entry 7) and so did the addition of MDHM to trimethylsilyl-
propyne (Table 4, entry 8).
Interestingly, the trimethylsilyl substituent proves to be a more
effective directing group than a t-Bu residue (Table 4, entry 8
vs entry 6) probably owing to its lower electronegativity, which
results in a greater polarization of the carbon-carbon triple
bond. This effect is particularly useful and can lead to excellent
levels of regiocontrol (Table 4, entry 10). Whereas hydrosily-
lation of 29 gave a 5:1 ratio in favor of the less substituted
isomer (Table 4, entry 9), we were curiously unable to effect
the addition of PhMe2SiH to but-2-yn-1-ol (Table 4, entry 11)
even when 1 mol % of catalyst 7 was employed.
FIGURE 4. Effect of the different reaction components on the catalyst
activation period. Curve A: Reaction of (IPr)Pt(AE) (7) and MDHM
for 3 h at 60 °C, followed by addition of 1-octyne at 20 °C. Curve B:
(IPr)Pt(AE) and 1-octyne for 3 h at 60 °C, followed by addition of
MDHM at 20 °C. Curve C: MDHM and 1-octyne, followed by addition
of (IPr)Pt(AE) at 20 °C. Reaction conditions: MDHM (6.0 mmol, 1
equiv), 1-octyne (6.0 mmol, 1 equiv), (IPr)Pt(AE) (0.1 mol %), THF
(1.0 M). The evolution of the reaction was followed by GC analysis.
catalytic species to initiate the hydrosilylation. The addition of
MDHM to alkenes, catalyzed by (IPr)Pt(DVDS) (6), displayed
a similar reaction profile albeit at 80 °C.45
To deconvolute the influence of each of the reaction
components on the kinetic profile of the hydrosilylation of
1-octyne by MDHM three separate experiments were performed.
Initially, we incubated the catalyst with MDHM at 60 °C for
3 h (Figure 4, eq A), then the catalyst and 1-octyne were allowed
to react for 3 h at 60 °C (Figure 3, eq B), and finally, as a
control experiment, the alkyne was treated with MDHM under
identical conditions (Figure 4, eq C). The three reactions were
then brought back to 20 °C and the third component was added
(the alkyne, the silane, and 7 in the first, second, and third
reaction, respectively), and the kinetics were measured. The
preactivation of (IPr)Pt(AE) with the silane led to a 6-fold
increase in the reaction rate and this treatment completely
suppressed the initial induction period (Figure 4, curve A vs
curve B). On the other hand, incubation of the platinum complex
with 1-octyne led to complete deactivation of the catalyst (Figure
4, curve C vs curve B). This observation reinforces our
proposition, made in the case of BnOP, that the allyl ether ligand
can be displaced by an alkyne to form the catalytically inactive
(IPr)Pt(alkyne)2 complex (Scheme 6), which can further shed
Catalyst Activation and Deactivation Pathways. The dif-
ficulties encountered in the hydrosilylation of BnOP (vide supra)
raised a number of questions on the modes of activation and
deactivation of the (IPr)Pt(AE) (7) catalyst. The kinetic profile
of our model reaction at room temperature revealed an induction
period of 40 min before hydrosilylation began (Figure 3). This
is the time necessary to generate a sufficient amount of active
(45) Berthon-Gelloz, G.; Brie`re, J. F.; Buisine, O.; Ste´rin, S.; Michaud, G.;
Mignani, G.; Tinant, B.; Declercq, J.-P.; Chapon, D.; Marko´, I. E. J. Organomet.
Chem. 2005, 690, 6156–6168.
(44) Molander, G. A.; Retsch, W. H. Organometallics 1995, 14, 4570–4575.
J. Org. Chem. Vol. 73, No. 11, 2008 4195

Berthon-Gelloz et al.
SCHEME 7. Proposed Catalytic Cycle and Deactivation Pathways for the Hydrosilylation of Alkynes Catalyzed by
(IPr)Pt(AE) 7
the NHC ligand to produce an unselective hydrosilylation
catalyst. Despite the generally accepted belief that carbene-metal
bonds are strong and do not dissociate easily, the loss of NHC
ligands has been reported in many cases.46
Investigations into the nature of the highly reactive intermedi-
ate formed by activation of 7 with a silane are currently
underway and will be reported in due course. Preliminary results
suggest the formation of a sensitive (IPr)Pt(SiR3)2 complex,
which is a bright yellow compound, reminiscent of the
characteristic yellow taint observed during the hydrosilylation
reaction (vide supra).
The information garnered on the reactivity of (IPr)Pt(AE)
enabled us to propose a reasonable catalytic cycle (Scheme 7).
The (IPr)Pt(AE) complex 7 can react with a silane to form the
[(IPr)Pt(H)(SiR3)] (A) intermediate, which subsequently under-
goes 1,2-migratory insertion of the alkyne to produce B. The
ꢀ-(E) isomer is favored over the R isomer because of the steric
repulsion encountered in the Pt alkenyl intermediates (B and
B′). After reductive elimination of the vinylsilanes, the [IPr-Pt]
(C) intermediate can either react with the silane, thus regenerat-
ing species A and thereby completing the catalytic cycle, or be
coordinated by two alkyne molecules to yield a bisalkyne
complex D, which is catalytically inactive.32 Platinum derivative
D can lose the IPr ligand to form the corresponding platinum
bisalkyne complex E, which, having now lost the bulky IPr
ligand, can perform the hydrosilylation reaction but with a low
regioselectivity. The (IPr)Pt(alkyne)2 (E) complex can also be
generated directly from (IPr)Pt(AE) and the alkyne by displace-
ment of the bidentate allyl ether ligand. Indeed, this was
observed when 1-octyne and BnOP (10) were incubated with
7. These catalyst deactivation pathways become predominant
when the acetylene has a strong coordinating ability (e.g.,
BnOP). To compensate for the loss of the catalyst, higher
loading must be used. Alternatively, the concentration of the
alkyne could be kept low.
loading (0.1 to 0.05 mol %), leads exclusively to the cis-addition
product containing the silyl substituent on the less hindered
terminus of the alkyne. Moreover, the regioselectivity of the
addition appears to be independent of the temperature. Inves-
tigations into the mechanistic cycles have enabled us to pinpoint
several deactivation pathways which are triggered by the alkyne.
Low regioselectivities and poor activities are due to a competing
secondary cycle in which a platinum species having lost the
stereodirecting IPr carbene ligand catalyzed the hydrosilylation.
The appearance of this undesired species is due either to the
strong coordinating ability of the alkyne or to the low reactivity
of the silane or to both. This situation can be remedied by
performing the reaction at higher temperatures (60 °C), by
increasing the catalyst loading, or by adding slowly the alkyne
to the catalyst/silane mixture.
On the basis of our previous study on the structure-activity
relationship of (NHC)Pt-based complexes32 and the present
study, our efforts are now being directed toward designing
NHCs that would lead to even higher selectivities for internal
alkynes.
Experimental Section
The general conditions employed for the hydrosilylation reactions
given in Tables 3 and 4 are described below. A detailed account
of the reaction conditions and the full characterization of products
can be found in the Supporting Information.
Synthesis of (IPr)Pt(AE) (7). To a freshly prepared solution of
Pt2(AE)3 (0.50 mmol of Pt, containing 10 equiv of excess AE) in
THF (5 mL, 0.1 M) was added the IPr ·HCl salt (331 mg, 0.75
mmol, 1.5 equiv) at rt, followed by t-BuOK (83 mg, 0.75 mmol,
1.5 equiv). The reaction was stirred at 20 °C until the reaction was
judged complete by TLC (Hex/Et2O 85:15), usually 16 h. The
mixture was filtered on a pad of Celigel (SiO2/Celite 1:1) and eluted
with Et2O. The crude complex was purified by flash chromatography
(Hex/Et2O 85:15) to yield 7 (256 mg, 75%). Rf 0.65 (SiO2, hexanes/
1
Et2O, 85/15, UV, heat, KMnO4). H NMR (500 MHz, CDCl3) δ
7.36 (t, 2H, J ) 7.7 Hz), 7.20 (s, 2H, JPt-H ) 9.0 Hz), 7.17 (d, 4H,
J ) 7.7 Hz), 3.95 (dd, 2H, J ) 12.0, J ) 3.5, JPt-H ) 34.4 Hz),
2.97 (h, 4H, J ) 6.7 Hz), 2.44 (dt, 2H, J ) 10.9 Hz, J ) 2.8 Hz,
JPt-H ) 30.0 Hz), 1.63 (tt, 2H, J ) 5.6 Hz, J ) 11.6 Hz), 1.32 (dd,
2H, J ) 1.0, J ) 8.3, JPt-H ) 61.6 Hz), 1.18 (d, 12H, J ) 6.9 Hz),
1.15 (d, 12H, J ) 6.9 Hz), 0.83 (dd, 2H, Jgem ) 1.0 Hz, J ) 10.7
Hz, JPt-H ) 60.0 Hz). 13C NMR (75 MHz, CDCl3) δ 187.5 (C),
Conclusions
In summary, we have designed a new NHC-platinum(0)
complex, (IPr)Pt(AE) 7, possessing enhanced activity and
displaying high regioselectivity in the hydrosilylation of terminal
and internal alkynes. The reaction, which requires low catalyst
145.9 (C), 136.8 (CH.), 129.1 (CH), 123.4 (CH), 123.2 (JPt-C
)
40.8 Hz, C), 69.2 (JPt-C ) 30.4 Hz, CH2), 47.8 (JPt-C ) 142.8 Hz,
CH), 32.1 (JPt-C ) 196.1 Hz, CH2), 28.4 (CH), 25.7 (CH3), 22.6
(46) Crudden, C. M.; Allen, D. P. Coord. Chem. ReV. 2004, 248, 2247–
2273.
4196 J. Org. Chem. Vol. 73, No. 11, 2008

Highly ꢀ-(E)-SelectiVe Hydrosilylation of Alkynes
(CH3). 195Pt NMR (107 MHz, CDCl3) -5574 ppm. IR (film, cm-1
)
transformations. Analytically pure samples were obtained by
bulb-to-bulb distillation.
3165 w, 3124 m, 3090 w, 3037 w, 2962 s, 2869 s, 2839 m, 1560 m,
1467 s, 1446 s, 1395 s, 1385 m, 1362 m, 1330 s, 1260 s, 1178 s,
1059 s, 929 s, 858 m, 801 s, 757 s, 697 m, 609 m. MS (ESI) m/z
681-680-599 [M - H]+, 625-624-623 [(IPr)Pt(CH2dCHs
CH2)]+, 599-598-597 [(IPr)Pt(CH3)]+. Anal. Calcd for
C33H46N2OPt: C, 58.13; H, 6.80; N, 4.11. Found: C, 58.19; H, 6.82;
N 4.04.
Acknowledgment. This paper is dedicated to Professor T.
Hayashi on the occasion of his 60th birthday. G.B.-G. acknowl-
edges Rhodia Corporate for a Ph.D. fellowship. We wish to
thank Umicore A.G. (Dr. R. Karch and V. Raab) for a generous
gift of platinum salts. Dr. B. Leroy is acknowledged for a gift
of compound 17. Dr. G. Binot is thanked for performing the
hydrosilylation on compound 29.
General Procedure for the Hydrosilylation of Alkynes
Catalyzed by (IPr)Pt(AE) (7). A mixture of silane (1.1 equiv)
and alkyne (1 equiv) was dissolved in the appropriate solvent
(1.0 M, solvent can be omitted) and brought to 60 °C. To this
mixture was added (IPr)Pt(AE) (7) (THF solution, 10 mg/mL,
0.1 mol %). The reaction was heated until judged complete by
GC analysis. The ꢀ-(E)/R ratio of the reaction were determined
by GC analysis. The solvents were removed and the residue was
filtered on a pad of silica (elution with hexane) to remove the
catalyst. The product obtained was pure enough for further
Supporting Information Available: Additional experimental
procedures and full characterization data of all compounds
reported. This material is available free of charge via the Internet
at http://pubs.acs.org.
JO800411E
J. Org. Chem. Vol. 73, No. 11, 2008 4197