Selective hydrosilylation of styrene using an in situ formed platinum(1,3-dimesityl-dihydroimidazol-2-ylidene) catalyst

Article abstract of DOI:10.1016/S0022-328X(03)00514-X

A highly active and selective in situ formed platinum(N-heterocyclic carbene) catalyst for the hydrosilylation of styrene with triethylsilane is described, which unlike all other known hydrosilylation catalysts, selectively yields hydrosilylation products, but (almost) no dehydrogenative silylation products.

Full text of DOI:10.1016/S0022-328X(03)00514-X

Journal of Organometallic Chemistry 679 (2003) 149Á152
/
www.elsevier.com/locate/jorganchem
Short communication
Selective hydrosilylation of styrene using an in situ formed
platinum(1,3-dimesityl-dihydroimidazol-2-ylidene) catalyst
Jeroen W. Sprengers a, Maayke J. Mars a, Marcel A. Duin a, Kingsley J. Cavell b,
Cornelis J. Elsevier a,*
a Institute of Molecular Chemistry, Organometallic and Coordination Chemistry, University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV
Amsterdam, The Netherlands
b Department of Chemistry, Cardiff University, P.O. Box 912, Cardiff, Wales CF11 3TB, UK
Received 7 May 2003; received in revised form 2 June 2003; accepted 2 June 2003
Abstract
A highly active and selective in situ formed platinum(N-heterocyclic carbene) catalyst for the hydrosilylation of styrene with
triethylsilane is described, which unlike all other known hydrosilylation catalysts, selectively yields hydrosilylation products, but
(almost) no dehydrogenative silylation products.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Platinum; Carbene; Hydrosilylation; Dehydrogenative silylation; Styrene
1. Introduction
should readily dissociate one or two moles of alkene,
thus forming a suitable species for activation of SiÃ
/H
Platinum-catalyzed hydrosilylation is an important
reaction in organosilicon chemistry, inter alia for the
production of silicon polymers [1]. The Speier catalyst
[2] (H2PtCl6/iPrOH) and Karstedt catalyst [3] (H2PtCl6/
divinyltetramethyldisiloxane) show very high activities
for this reaction, but exhibit low stabilities: rapid
formation of inactive, colloidal platinum takes place.
Development of new hydrosilylation catalysts is there-
fore a continuing challenge and indeed several other
bonds and hence be active in catalytic hydrosilylation. A
recent paper by Marko´ et al. [9], in which some
platinum(0)-carbene complexes (2) have been reported
for the efficient hydrosilylation of 1-octene and
1,1,1,3,5,5,5-heptamethyltrisiloxane, indeed shows that
this reasoning is correct.
active platinum-complexes have been reported [1,4Á6].
/
The use of N-heterocyclic carbenes (NHC) as ligands
in homogeneous catalysis and organometallic chemistry
is rapidly expanding at the moment [7]. Recently, we
synthesized [Pt(IMes)(dimethylfumarate)2] (1a) (IMesꢀ
1,3-dimesityl-imidazol-2-ylidene) and [Pt(SIMes)(di-
methyl fumarate)2] (1b) (SIMesꢀ1,3-dimesityl-dihy-
droimidazol-2-ylidene) which are capable of activating
the CÃH bond of imidazolium salts at ambient tem-
perature [8]. We reasoned that species like 1a and 1b
/
Here, we report the use of new platinum(0) complexes
derived from bulky unsaturated and saturated carbenes
in the hydrosilylation of styrene with triethylsilane,
which proceeds with unprecedented high selectivity,
yielding exclusively products I and II. The platinum-
catalyzed hydrosilylation of styrene yielding the b- and
a-silylated products I and II is normally accompanied
by appreciable amounts of III and IV due to dehydro-
genative silylation.
/
/
* Corresponding author. Tel.: ꢁ
6456.
E-mail address: elsevier@science.uva.nl (C.J. Elsevier).
/
31-20-525-5653; fax: ꢁ31-20-525-
/
0022-328X/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00514-X

150
J.W. Sprengers et al. / Journal of Organometallic Chemistry 679 (2003) 149Á152
/
Table 1
Hydrosilylation of styrene with triethylsilane
a
Entry Catalyst
Yield
(%)
TOF b
Selectivity (%)
I:II:III
(hꢂ1
)
1
2
3
4
5
6
4ꢁ
4ꢁ
4ꢁ
4ꢁ
4ꢁ
4ꢁ
/
ImesHCl c,d
IMesHCl c,e
IMesHCl c,f
I-t BuHCl c,f
SIMesHCl c,f
SI-
18.9
8.1
9.8
85.2:3.2:11.0
76.6:19.9:2.8
78.3:15.9:5.8
69.7:3.5:26.8
82.5:17.2:0.3
83.2:16.6:0.2
/
26.7
31.2
7.2
99
/
12.7
2.1
/
/
ꢀ
ꢀ/
/
53.4
80.2
2. Results and discussion
/
99
MesHCl c,f,g
SI-t BuHCl c,f
SI-
4.5
0
2.3
0
92.2:0.7:7.1
Initially, we applied [Pt(IMes)2] (3) [10] as a catalyst,
but a very low activity was observed even in refluxing
toluene. Apparently, two carbenes coordinated to the
platinum center results in a complex that is too stable
for effective catalytic hydrosilylation. We reasoned that
a platinum(mono-carbene) complex containing two
labile ligands, such as alkenes, should be a suitable
precursor to a platinum(carbene) species exhibiting
good catalytic performance. Therefore attempts were
made to synthesize [Pt(IMes)(norbornene)2] by stirring
[Pt(norbornene)3] (4) with one equivalent of IMesHCl
7
8
4ꢁ
/
4ꢁ
/
Á/
DippHCl c,f
9
1a
1b
3
5 h
5
5 g
18.0
76.2
11.1
41.6
2
64.7:14.9:20.4
83.1:16.8:0.1
35.3:11.6:53.1
78.1:20.1:1.7
79.6:19.9:0.5
79.5:19.7:0.8
10
11
12
13
14
B/2
4.9
5.2
9.1
12.4
24.6
34.6
Triethylsilane/styrene ratio 1.0, 1 mol% of catalyst, toluene, 100 8C.
Total GC-Yield after 3 h using n-decane as internal standard.
a
b
Turn-over frequency, determined at 25% conversion, defined as
moles of styrene converted per mole of catalyst per hour.
Complex 4 stirred with one equivalent of the imidazolium-salt and
two equivalents of KO-t Bu for a period of time.
No complex formation time.
(IMesHClꢀ1,3-dimesityl-imidazolium chloride) and
/
c
1.5 equivalents of NaH in THF at room temperature
1
for 16 h. A stable white solid was obtained, H-NMR
and 195Pt-NMR revealed that it was a mixture of two
complexes, which could not be separated. Two platinum
d
e
Complex formation time of 30 min.
Complex formation time of 60 min.
f
g
Triethylsilane/styrene ratio 3.0.
Triethylsilane/styrene ratio 0.33.
resonances, at ꢂ
/
3821 and ꢂ3940 ppm, were observed
/
h
1
and the H-NMR spectrum showed a carbene/norbor-
nene ratio of 1. Addition of one equivalent of maleic
anhydride to this solid gave the white pure complex
[Pt(IMes)(maleic anhydride)(2-norbornene)] (5) in 74%
yield. The synthesis of the analogous complex with the
saturated carbene ligand, starting from SIMesHCl, was
also attempted, but was not successful.
as catalysts. Results of the hydrosilylation of styrene
with triethylsilane in toluene at 100 8C using these
catalysts have been compiled in Table 1.
No induction period was observed for any of the
catalytic hydrosilylation reactions and plots of conver-
Since the activity of platinum(0)-complexes strongly
depends on the nature of the coordinated alkenes, [4]
catalysis by in situ formed catalysts derived from
[Pt(norbornene)3] (4) and NHC is interesting. Thus,
the Pt(NHC) catalyst was prepared by stirring an
imidazolium-salt with two equivalents of KO-tBu and
one equivalent of 4, after which catalysis was started by
addition of styrene and triethylsilane. The isolated
platinum-compounds 1a, 1b, 3 and 5 were also evaluated
sion against time gave straight lines, indicating that all
catalysts were stable under the reaction conditions.
Firstly, an in situ formed catalyst from IMesHCl and
4 was investigated. Using pre-set complex-formation
times of 0, 30 and 60 min (entries 1Á3) it became clear
/
that the yields and turn-over frequencies (TOFs) steadily
increase with longer complex formation periods. Ap-
parently, a certain time is required for the formation of
the catalytically active complex. Therefore, for all

J.W. Sprengers et al. / Journal of Organometallic Chemistry 679 (2003) 149Á
/
152
151
further catalytic experiments involving in situ formation
of the catalyst from these and comparable precursors, a
complex-formation time of 60 min was used. If the
formation time for the complex was omitted when the
catalyst was formed from IMesHCl and 4, the selectivity
observed was similar to that observed using 4 alone. By
applying pre-set periods of 30 (entry 2) and 60 min
(entry 3), indeed less dehydrogenative silylation was
observed: 2.8 and 5.8%, respectively. However, these
experiments yielded an increased amount of the
branched hydrosilylation product (II), 19.9 and 15.9%,
respectively. When the imidazolium salt I-tBuHCl (entry
4) was used, which yields a carbene ligand that was also
TOF of only 2 hꢂ1 was obtained together with a high
degree of dehydrogenative silylation (53.1%).
The higher reactivity of platinum complexes derived
from SIMesHCl relative to its unsaturated analogue can
be explained by the difference in s-donor strength
between these two ligands. Although all NHC are
almost pure s-donors, saturated carbene ligands exert
an even larger s-donor strength. Presumably, the
oxidative addition of triethylsilane to the platinum(0)
is rate-determining, which step is favored by a higher
electron density on the platinum center. Studying
complex 5 using different triethylsilane/styrene ratios
demonstrated this once more. A higher activity was
obtained by using a larger triethylsilane/styrene ratio
used by Marko´ et al. [9], a very low activity (TOFꢀ2.1
/
hꢂ1) and an increased amount of dehydrogenative
silylation (26.8%) was obtained.
Interestingly, when the saturated imidazolium chlo-
(increasing from 0.33 to 3; entries 12Á14), indicating a
/
positive order on the rate in triethylsilane. Probably,
styrene acts as an inhibitor, thereby blocking the
oxidative addition of triethylsilane.
ride, SIMesHCl, (SIMesHClꢀ1,3-dimesityl-4,5-dihy-
/
dro-imidazolium chloride) was employed (entry 5)
instead of the unsaturated analogues used so far, a
Summarizing, we have shown that the in situ gener-
ated platinum(1,3-dimesityl-dihydroimidazol-2-ylidene)
catalyst is by far the most efficient Pt(carbene) catalyst
for the hydrosilylation of styrene with triethylsilane.
High yields of product and almost 100% selectivity
towards hydrosilane addition products, and the virtual
absence of dehydrogenative silylation make it a very
effective catalytic system. Further studies aimed at
elucidating details of this hydrosilylation reaction and
the platinum compounds involved are in progress.
very high activity (TOFꢀ
53.4 hꢂ1) was observed.
/
Significantly, but completely unexpectedly, dehydro-
genative silylation was virtually absent (only 0.3%) in
this case, whereas it is known to occur in all reported
cases [11Á13]. Furthermore, this side-reaction is strongly
/
influenced by the ratio triethylsilane/styrene with those
catalysts. However, no change in selectivity was ob-
served with our catalyst when using a triethylsilane/
styrene ratio of 3 (entry 6), but the TOF (80.2 hꢂ1
)
significantly increased. Similar to what is observed with
IMesHCl, a relatively large amount (about 17%) of the
regioisomer II is formed.
3. Experimental
Using SI-tBuHCl (entry 7) yielded a slow catalyst
however the selectivity of the reaction changed: a
considerable amount III was formed, 7.1%, but product
II was virtually absent. When employing SIDippHCl,
3.1. Synthesis of [Pt(IMes)(maleic anhydride) (2-
norbornene)]
NaH (31 mg, 0.78 mmol) and 1,3-dimesityl-imidazo-
lium chloride (171 mg, 0.50 mmol) were suspended in
tetrahydrofuran (10 ml). Then 4 (240 mg, 0.50 mmol)
was added to the mixture which was stirred at room
temperature for 18 h. The solution was filtered and
maleic anhydride (46.5 mg, 0.47 mmol) was added and
the solvent was evaporated after 10 min of stirring at
room temperature. The solid compound was washed
(SIDippHClꢀ1,3-(2,6-diisopropylphenyl)-4,5-dihydro-
/
imidazolium chloride) no catalytic activity was observed
(entry 8). It would appear that the isopropyl-substitu-
ents are too stericly demanding and therefore interfere
with either the addition of the hydrosilane or the
addition of styrene to the platinum center.
The high activity and selectivity of the platinum
compounds derived from SIMesHCl was confirmed by
testing the pre-formed platinum(0) complexes 1a (entry
9) and 1b (entry 10). Indeed, complex 1b, with the
saturated carbene ligand, was more active and selective
than 1a. However, overall lower yields were obtained
when compared to the in situ prepared catalysts. The
selectivity obtained with complex 1b is the same as the in
situ generated catalyst (entry 5). This is not the case for
complex 1a, where more dehydrogenative silylation
(20%) was observed for the pre-formed compound.
These complexes give rise to quite effective catalysts
when compared to [Pt(IMes)2] (entry 11). As mentioned,
the known complex [Pt(IMes)2] is very inefficient: a
with ether (3ꢃ2 ml) and dried in vacuo. A light yellow
/
powder was obtained. Yield: 249 mg, 0.36 mmol, 74%;
1H-NMR (499.8 MHz, benzene-d6, d (ppm)): 6.69 (s,
4
2H), 6.47 (s, 2H), 6.14 (s, JPtH 9.4 Hz, 2H), 3.34 (d,
3JHH 4.4 Hz, 2JPtH
ꢀ
49.6 Hz, 1H), 3.17 (d, 3JHH 5.4 Hz,
/
2JPtH 66.9 Hz, 1H), 2.87 (d, JHH 4.4 Hz, JPtH
ꢀ49.3
/
3
2
3
2
Hz, 1H), 2.76 (d, JHH 5.4 Hz, JPtH 65.5 Hz, 1H), 2.70
(s, 1H), 2.23 (s, 1H), 1.93 (s, 18H), 1.39 (m, 1H), 1.23 (m,
3
1H), 0.93 (m, 1H), 0.84 (m, 1H), 0.39 (d, JHH 8.6 Hz,
3
1H), 0.05 (d, JHH 8.8 Hz, 1H). 13C-NMR (125.7 MHz,
1
benzene-d6, d(ppm)): 175.9 (NCN, JPtC 1425.7 Hz),
175.3 (CÄ
/
O, 2JPtC 32.9 Hz), 171.2 (CÄ
/
O, 2JPtC 41.0 Hz),
139.2, 128.3, 71.9 (CHÄ
/
CH, 1JPtC 191.7 Hz), 67.1 (CHÄ
/

152
J.W. Sprengers et al. / Journal of Organometallic Chemistry 679 (2003) 149Á
/
152
1
1
CH, JPtC
CH, JPtC 199.8 Hz), 43.7, 43.2, 41.9 (CHÄ
/
ject number 2000-14), which is gratefully acknowledged
for financial support.
207.5 Hz), 41.2 (3JPtC 45.6 Hz), 40.5 (CHÄ
/
CH, JPtC
148.6 Hz), 28.4 (3JPtC 45.3 Hz), 28.2 (3JPtC 44.5 Hz),
20.9, 17.8. 195Pt-NMR (85.6 MHz, benzene-d6, d
1
(ppm)): ꢂ5291. Anal. Found: C 55.48, H 5.21, N 4.02.
/
Calc.: C 55.56, H 5.25, N 4.05%.
References
3.2. Catalytic hydrosilylation of styrene
[1] B. Marciniec, Comprehensive Handbook on Hydrosilylation,
Pergamon Press, London, 1992.
A Schlenk tube equipped with septum, condenser
cooler and stirring bar was charged with 0.05 mmol of
the relevant complex (Table 1) with or without one
equivalent of an imidazolium-salt and two equivalents
of KO-tBu. The Schlenk tube was evacuated and filled
with nitrogen three times and 20 ml toluene. Then, 5.0
or 15.0 mmol of styrene, 3.0 mmol of n-decane and 5.0
or 15.0 mmol of triethylsilane were added in quick
succession via syringes. The reactor was immediately
immersed in an oil bath, which was kept at 100 8C.
Samples were taken periodically for GC analysis.
[2] J.L. Speier, J.A. Webster, G.H. Barnes, J. Am. Chem. Soc. 79
(1957) 974.
[3] B.D. Karstedt (General Electric Co.) USP 226928 (1972).
[4] P. Steffanut, J.A. Osborn, A. DeCain, J. Fisher, Chem. Eur. J. 4
(1998) 2008.
[5] K. Roy, R.B. Taylor, J. Am. Chem. Soc. 124 (2002) 9510.
[6] M.N. Jagadeesh, W. Thiel, J. Kohler, A. Fehn, Organometallics
¨
21 (2002) 2076.
[7] W.A. Herrmann, Angew. Chem. Int. Ed. Engl. 41 (2002)
1290.
[8] M.A. Duin, N.D. Clement, K.J. Cavell, C.J. Elsevier, Chem.
Commun. (2003) 400.
[9] E. Marko´, S. Ste´rin, O. Buisine, G. Mignani, P. Branlard, B.
Tinant, J.-P. Declercq, Science 298 (2002) 204.
[10] J. Arduengo, III, S.F. Gamper, J.C. Calabrese, F. Davidson, J.
Am. Chem. Soc. 116 (1994) 4391.
Acknowledgements
[11] Hopf, K.H. Dotz, J. Mol. Catal. A 164 (2000) 191.
¨
[12] W. Caseri, P.S. Pregosin, J. Organomet. Chem. 356 (1988) 259.
[13] J.W. Sprengers, M. de Greef, M.A. Duin, C.J. Elsevier, Eur. J.
Inorg. Chem., 2003, in press.
The research has been performed within the Dutch
National Research School Combination Catalysis (pro-