Chemoselective catalytic hydrosilylation of nitriles

Article abstract of DOI:10.1002/anie.201003069

Well-behaved: The ruthenium complex [Cp(iPr₃P)Ru(NCCH ₃)₂]+ (Cp=cyclopentadienyl) catalyzes monohydrosilylation of nitriles (see scheme) with unprecedented tolerance of most common functional groups. The catalyst is easily synthesized starting from commercially available compounds, is air-stable, recyclable, and works under solvent-free conditions.

Full text of DOI:10.1002/anie.201003069

Angewandte
Chemie
DOI: 10.1002/anie.201003069
Silyl Compounds
Chemoselective Catalytic Hydrosilylation of Nitriles**
Dmitry V. Gutsulyak and Georgii I. Nikonov*
Dedicated to Professor Lyudmila G. Kuzmina on the occasion of her 65th birthday
Catalytic hydrosilylation of unsaturated substrates has
emerged over recent decades as a powerful industrial and
laboratory methodology for the preparation of a wide range
of organosilicon products.[1] In the case of ketones and imines,
it also serves as a convenient reduction method in that it
provides protected alcohols and amines in one step.[2]
Whereas diverse procedures have been developed for the
Other highlights of this catalytic system are that it can operate
under solvent-free conditions and that the catalyst is recycla-
ble.
We have recently reported that cationic Ru complexes
[Cp(R3P)Ru(NCCH3)2]+ (1; Cp = cyclopentadienyl) catalyze
a variety of hydrosilylation reactions, showing moderate to
excellent activity.[15] Rewardingly, we found that complex
[Cp(iPr3P)Ru(NCCH3)2]BAF (1a; BAF = [B(C6F5)4]À) also
catalyzed the hydrosilylation of nitriles better than it does the
[1]
[3]
[4]
=
=
=
hydrosilylation of C X (X = C, N, O), P O, S O, and
[1,5]
ꢀ
À
ꢀ
C C bonds,
Si H addition to the N C triple bond remains
a great synthetic challenge.[1h,6] Hydrosilylation of nitriles also
presents a difficult chemoselectivity problem in that the
hydrosilylation of C O and C C bonds (Table 1). Simple
alkyl- and aryl-substituted nitriles RCN (Table 1, entries 1–4)
are easily converted at room temperature into the corre-
sponding N-silylimines. The reaction times appear to increase
with the size of group R. Hydrosilylation of tBuCN is still
selective, but requires slightly increased temperature (508C)
and a longer reaction time (Table 1, entry 3). The CN group in
the homoallylic position is selectively reduced in the presence
=
=
=
products of monoaddition, N-silylaldimines R’3Si-N CHR,
are usually much more reactive than nitriles, so that the
reaction proceeds further to give the disilylamines
(R’3Si)2NCH2R.[6c,7] Since the original report by Calas et al.[8]
on the ZnCl2-catalyzed condensation of HSiEt3 with PhCN to
=
give the imine PhHC NSiEt3 in moderate yield (54%, at 140–
=
1508C), only a few catalytic monohydrosilylations of nitriles
have been published.[9,10] These methods either require drastic
reactions conditions or are very slow and give low yields, and
in no case was the selectivity towards functional groups
of C C double bond (Table 1, entry 5). In contrast, the
hydrosilylation of a conjugated alkenyl nitrile required much
longer time and does not reach completion even after 24 h
(Table 1, entry 6), suggesting possible catalyst poisoning by
the chelating azabutadiene product. Previous reports on the
hydrosilylation of unsaturated nitriles primarily showed
selective Si H addition to the C C double bond
only a few examples of double hydrosilylation of the cyano
group.[7b] Attempted hydrosilylation of a nonconjugated
alkynyl nitrile (Table 1, entry 7) resulted in a very sluggish
Si H addition to the C C triple bond, in accord with the high
propensity of half-sandwich complexes of Ru to activate
alkynes.[17] Electron-poor aryl nitriles containing the usually
reactive keto-, aldo-, nitro-, and ester functionalities in the
aryl substituent are hydrosilylated exclusively on the cyano
group (Table 1, entries 8–11), although the reaction times are
significantly increased relative to that of benzonitrile
(Table 1, entry 4). In contrast, HSiMe2Ph adds to the elec-
tron-rich p-cyanoanisole at the same rate (100% after 1 h in
[D6]acetone; Table 1, entry 12) as to PhCN (Table 1, entry 4).
Finally, hydrosilylation of 3-cyanopyridine occurs selectively
at the cyano group, but a long reaction time is required
(Table 1, entry 13).
established. A handful of examples of stoichiometric mono-
[11]
À
Si H additions to nitriles have been published.
[6b,16]
À
=
Given the importance of silylimines in medicinal chemis-
with
=
try and organic synthesis, the selective formation of R’3SiN
CHR from nitriles would be a significant addition to the
current repertoire of synthetic chemistry.[9,12] This process also
presents an attractive alternative to the current synthetic
methods of reduction of nitriles to imines and aldehydes
based on the use of expensive and often pyrophoric aluminum
or boron hydrides, such as diisobutylaluminum hydride
(DIBAL-H).[13] Also, silylaldimines have recently been
À
ꢀ
À
found to be useful partners in N C coupling with aryl halides
to make substituted aldimines, which paves the way to a
variety of secondary amines.[14]
Herein we report a convenient method for selective
mono(hydrosilylation) of nitriles that occurs under very mild
conditions and shows excellent tolerance to most common
functional groups. Moreover, simply increasing the silane load
allowed us to achieve the complete reduction to disilylamines.
The exceptional tolerance of this nitrile hydrosilylation to
the presence of the carbonyl functionality was also manifested
by performing the catalysis in acetone as a solvent, which
significantly accelerates the reaction rate. Thus, the full
conversion of benzonitrile into the silylated imine was four
times faster in acetone than in chloroform. Hydrosilylation of
acetonitrile is also faster in acetone, but is accompanied by
some loss of selectivity. Moreover, after 6 h most of the initial
[*] Dipl.-Chem. D. V. Gutsulyak, Dr. G. I. Nikonov
Chemistry Department, Brock University
500 Glenridge Ave., St. Catharines, ON L2S 3A1 (Canada)
Fax: (+1)905-682-9020
E-mail: gnikonov@brocku.ca
[**] This work was supported by the NSERC and OPIC. We are grateful to
the CFI/OIT for a generous equipment grant.
=
product, the imine CH3CH NSiMe2Ph, was converted into a
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201003069.
mixture of compounds, the major component of which was an
Angew. Chem. Int. Ed. 2010, 49, 7553 –7556
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7553

Communications
Table 1: Catalytic hydrosilylation of nitriles with HSiMe2Ph.[a]
Entry
Substrate
TOF [hÀ1 [b]
]
Conv. (time)[c]
Products
=
1
2
3
4
CH3CN
iPrCN
tBuCN
PhCN
75 (100%)
16.6 (50%)
0.04 (14%)
6.3 (50%)
100% (20 min)[d]
100% (2 h)
CH3CH NSiMe2Ph (90%)
=
iPrCH NSiMe2Ph
=
14% (24 h)
tBuCH NSiMe2Ph
100% (4 h)[e]
PhCH NSiMe2Ph
=
5
2.5 (100%)
100% (2.5 h)
=
=
=
=
6
7
CH2 CHCN
0.3 (30%)
0.06 (40%)
30% (24 h)
40% (168 h)
CH2 CHCH NSiMe2Ph
ꢀ
HC C(CH2)3CN
Me2PhSiCH CH(CH2)3CN
8
4.4 (50%)
100% (8 h)
9
4.6 (50%)
19 (50%)
0.5 (50%)
100% (14 h)
100% (4 h)
10
11
100% (48 h)[f]
12
13
25 (100%)
6.9 (50%)
100% (1 h)[d]
68% (14 h)
[a] In a general procedure, 4–5 mol% of 1 was added to a solution of substrate and silane in CDCl3. [b] Calculated at the conversion shown in
parentheses. [c] Calculated from 1H NMR data. [d] Reaction in [D6]acetone. [e] 100% after 1 h in [D6]acetone. [f] Reaction in CD2Cl2.
= À
unusual
coupling
product
CH3CH N CH(CH3)N-
resulted in only insignificant increase of reaction times,
without any drastic effect on the selectivity of the reaction.
Finally, the catalyst can be easily generated in the reaction
flask by treating the commercially available and shelf-
storable precursor [CpRu(NCCH3)3]+AÀ (AÀ = [PF6]À or
[BF4]À) with phosphine prior to catalysis.
Full reduction of nonfunctionalized nitriles to disilyl-
amines can easily be achieved with the same catalyst
(Table 2). Thus, bis(silyl)benzylamine is produced in the
reaction of benzonitrile with two equivalents of silane
(Table 2, entry 1).[20] For comparison, there was no hydro-
(SiMe2Ph)2. It is tempting to speculate that this compound
À
=
results from the so far unknown N Si addition across the C N
bond, but the exact mechanism is not known yet.
Importantly, we also discovered that the hydrosilylation of
liquid substrates can be carried out under solvent-free
conditions with reduced loadings of the catalyst (< 1%
mol). Taking advantage of the low solubility of the cationic
catalyst in nonpolar solvents, the hydrosilylation products can
be extracted with hexane or ether and the catalyst can easily
be recovered and used again.[18]
=
Changing the phosphine ligand in the ruthenium complex
or using a different silane has a drastic effect on the catalytic
activity. The best results were achieved for the system iPr3P/
HSiMe2Ph. All other combinations were much less active or
completely inactive.[19] Since the precatalyst is a cationic
complex, we also studied the influence of counteranion.
Rewardingly, the substitution of the expensive BAF anion for
the more readily available [PF6]À or [BF4]À counteranions
silylation of the related imine PhMeC NPh under the same
reaction conditions. p-Cyanoanisole is converted cleanly into
the bis(silylated)amine after 48 h (Table 1, entry 2), but the
reaction stops at the imine stage for the electron-poor nitrile
meta-NO2(C6H4)CN (Table 1, entry 3). Functionalized nitriles
gave a mixture of products because of loss of chemoselectivity
at the second stage of hydrosilylation (e.g. Table 1, entry 4).
Finally, double hydrosilylation of an enolizable alkylnitrile
7554
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7553 –7556

Angewandte
Chemie
Table 2: Hydrosilylation of nitriles with two equivalents of HSiMe2Ph catalyzed by 5 mol% 1a.[a]
Experimental Section
Catalytic hydrosilylation of benzoni-
trile: Method 1: Complex 1a (7 mg,
Entry
Substrate
Conv. (time)
Products
1
2
3
4
PhCN
100% (66 h)
100% (48 h)
100% (48 h)
100% (48 h)
PhCH2N(SiMe2Ph)2
MeO(C6H4)CH2N(SiMe2Ph)2
4% mol) was added to a solution of
benzonitrile (16.6 mL, 0.16 mmol) and
HSiMe2Ph (26 mL, 0.17 mmol) in
CDCl3. After 4 h at ambient temper-
ature, 100% conversion of the starting
nitrile to the N-silylimine was achieved.
MeO(C6H4)CN
m-NO2(C6H4)CN
p-CH3(O)C(C6H4)CN
=
m-NO2(C6H4)CH N(SiMe2Ph)
=
p-CH2 C(OSiMe2Ph)(C6H4)CH2N(SiMe2Ph)2
(40% in the mixture)
5
iPrCN
100% (24 h)
(CH3)2CHCH2N(SiMe2Ph)2 (43%)
=
(CH3)2C CHN(SiMe2Ph)2 (57%)
Method 2: Complex [Cp(iPr3P)Ru-
(NCCH3)2][PF6] (1b; 8.6 mg, 5 mol%)
was added to a solution of HSiMe2Ph
[a] See Experimental Section for details.
(47 mL, 0.30 mmol) and benzonitrile
(30 mL, 0.29 mmol) in [D6]acetone
was accompanied by the concomitant formation of an
enamine (Table 1, entry 5).
(0.6 mL). After 1 h at ambient temperature, 100% conversion of
the starting nitrile to the N-silylimine was achieved.
Method 3 (reaction without solvent): Complex 1b (106 mg,
1 mol%) was added to a mixture of HSiMe2Ph (3.0 mL, 19 mmol)
and benzonitrile (2.0 mL, 19 mmol). After 24 h at ambient temper-
ature, all benzonitrile was cleanly converted into the N-silylimine.
The product was separated from the catalyst by extraction with
hexane.
Catalytic hydrosilylation of isobutyronitrile (on the bench):
Catalyst 1b (0.054 g, 1.5% mol) was loaded in air into a 100 mL
round-bottom flask. The flask was purged with nitrogen and charged
with 30 mL of CH2Cl2. Isobutyronitrile (0.6 mL, 6.7 mmol) and then
HSiMe2Ph (1.1 mL, 7.1 mmol) were added. The reaction mixture was
stirred for 3 h at ambient temperature. Then hexane (30 mL) was
added, and the yellow solution became cloudy. The volume was
reduced to 30 mL resulting in precipitation of the catalyst as a brown
oil (contaminated with siloxanes). The yellow solution was decanted
from the precipitate and distilled under reduced pressure. Yield of the
imine: 1.2 g (86%). 1H NMR (CDCl3): d = 8.24 (d, 3J(H-H) = 4.4 Hz,
1, CH = N), 7.55–7.50 (m, 2, Ph), 7.40–7.35 (m, 3, Ph), 2.37 (m, 1, CH),
1.07 (d, 3J(H-H) = 6.6 Hz, 6, CH3), 0.44 ppm (s, 6, SiMe).
Our previous mechanistic studies on the Ru-catalyzed
hydrosilylation of carbonyls revealed a mechanism including
the intermediate formation of a silane s-complex 2, which
then undergoes nucleophilic abstraction of the silylium ion by
the substrate L to form a neutral ruthenium hydride complex
3 (Scheme 1).[15] Although several complexes of type 2 have
been isolated and characterized, our attempts at preparing the
labile compound 3 have been so far unsuccessful.[21] We
tentatively assign a similar mechanism for the hydrosilylation
of nitriles.[22]
Received: May 20, 2010
Revised: June 29, 2010
Keywords: chemoselectivity · homogeneous catalysis ·
.
hydrosilylation · ruthenium
[1] a) Hydrosilylation, Advances in Silicon Science (Ed.: B. Marci-
niec), Springer, Berlin, 2009; b) A. K. Roy, Adv. Organomet.
Chem. 2008, 55, 1; c) S. E. Gibson, M. Rudd, Adv. Synth. Catal.
2007, 349, 781; d) B. Marciniec, Appl. Organomet. Chem. 2000,
14, 527; e) R. Noyori, Asymmetric Catalysis in Organic Synthesis,
Wiley, New York, 1994; f) I. Ojima, Catalytic Asymmetric
Synthesis, VCH, New York, 1993; g) H. Brunner, W. Zettimeier,
Handbook of Enantioselective Catalysis with Transition Metal
Compounds, VCH, Weinheim, 1993; h) I. Ojima in The Chemis-
try of Organic Silicon Compounds (Eds.: S. Patai, Z. Rappo-
port), Wiley, New York, 1989, chap. 25.
Scheme 1. Proposed mechanism for the Ru-catalyzed hydrosilylation. L
is a carbonyl compound or nitrile, and the square represents an empty
coordination site.
Attempts to extend this methodology to an open-flask
setup were hampered by the sensitivity of the products to air.
However, thorough exclusion of water is not required, as the
silane reacts quickly with trace water in the presence of 1a to
give siloxane. The reactions can therefore be conveniently run
on the bench in closed flasks that have been preflashed with
inert gas.
[2] a) O. Riant, N. Mostefai, J. Courmarcel, Synthesis 2004, 2943;
b) J.-F. Carpentier, V. Bette, Curr. Org. Chem. 2002, 6, 913.
[3] C. Petit, A. Favre-Reguilon, B. Albela, L. Bonneviot, G.
Mignani, M. Lemaire, Organometallics 2009, 28, 6379.
[4] A. C. Fernandes, C. C. Rom¼o, Tetrahedron 2006, 62, 9650.
[5] a) C. H. Jun, R. H. Crabtree, J. Organomet. Chem. 1993, 447,
177; b) M. P. Doyle, K. G. High, C. L. Nesloney, T. W. Clayton, J.
Lin, Organometallics 1991, 10, 1225; c) T. Takahashi, F. Bao, G.
Gao, M. Ogasawara, Org. Lett. 2003, 5, 3479; d) B. M. Trost,
Z. T. Ball, J. Am. Chem. Soc. 2005, 127, 17644; e) M. V. Jimꢀnez,
J. J. Pꢀrez-Torrente, M. I. Bartolomꢀ, V. Gierz, F. J. Lahoz, L. A.
Oro, Organometallics 2008, 27, 224.
In summary, we discovered a very efficient catalytic
hydrosilylation of nitriles that shows unprecedented chemo-
selectivity and tolerates most common functional groups. The
catalyst is air-stable and can easily be assembled prior to use
starting from commercially available compounds. Moreover,
the catalyst is recyclable, thus making it very attractable for
practicable applications.
Angew. Chem. Int. Ed. 2010, 49, 7553 –7556
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7555

Communications
[6] a) R. Calas, Pure Appl. Chem. 1966, 13, 61; b) A. J. Chalk,
2176; e) M. Ochiai, H. Hashimoto, H. Tobita, Angew. Chem.
2007, 119, 8340; Angew. Chem. Int. Ed. 2007, 46, 8192.
[12] M. Panunzio, P. Zarantonello, Org. Process Res. Dev. 1998, 2, 49.
[13] M. Hudlicky, Reductions in Organic Chemistry, ACS, New York,
1996.
J. Organomet. Chem. 1970, 21, 207; c) R. J. P. Corriu, J. J. E.
Moreau, M. Pataud-Sat, J. Organomet. Chem. 1982, 228, 301;
d) B. Marciniec, Comprehensive Handbook on Hydrosilylation,
Pergamon, Oxford, 1992.
[7] a) T. Murai, T. Sakane, S. Kato, J. Org. Chem. 1990, 55, 449 – 453;
b) T. Murai, T. Sakane, S. Kato, Tetrahedron Lett. 1985, 26, 5145 –
5148; c) A. M. Caporusso, N. Panziera, P. Petrici, E. Pitzalis, P.
Salvadori, G. Vitulli, G. Martra, J. Mol. Catal. A 1999, 150, 275 –
285.
[14] J. Barluenga, F. Aznar, C. Valdꢀs, Angew. Chem. 2004, 116, 347;
Angew. Chem. Int. Ed. 2004, 43, 343.
[15] D. V. Gutsulyak, S. F. Vyboishchikov, G. I. Nikonov, J. Am.
Chem. Soc. 2010, 132, 5950.
[16] B. A. Bluestein, J. Am. Chem. Soc. 1961, 83, 1000.
[17] B. M. Trost, M. U. Frederiksen, M. T. Rudd, Angew. Chem. 2005,
117, 6788; Angew. Chem. Int. Ed. 2005, 44, 6630.
[8] R. Calas, E. Frainnet, A. Bazouin, C. R. Hebd. Seances Acad. Sci.
1961, 252, 420 – 422.
[9] T. Fuchigami, I. Igarashi, Jpn Patent Appl. JP11228579, 1999.
[10] Catalytic monohydrosilylation of nitriles has been reported but
with a low product yield or a slow reaction rate: a) A. Y.
Khalimon, R. Simionescu, L. G. Kuzmina, J. A. K. Howard, G. I.
Nikonov, Angew. Chem. 2008, 120, 7815; Angew. Chem. Int. Ed.
2008, 47, 7701; b) E. Peterson, A. Y. Khalimon, R. Simionescu,
L. G. Kuzmina, J. A. K. Howard, G. I. Nikonov, J. Am. Chem.
Soc. 2009, 131, 908.
[11] a) J. Kim, Y. Kang, J. Lee, Y. K. Kong, M. S. Gong, S. O. Kang, J.
Ko, Organometallics 2001, 20, 937; b) M. Tanabe, K. Osakada,
Organometallics 2001, 20, 2118; c) H. Hashimoto, I. Aratani, C.
Kabuto, M. Kira, Organometallics 2003, 22, 2199; d) T. Wata-
nabe, H. Hashimoto, H. Tobita, J. Am. Chem. Soc. 2007, 128,
[18] The importance of this strategy has been highlighted in: V. K.
Dioumaev, M. Bullock, Nature 2000, 424, 530.
[19] See the Supporting Information for details.
[20] When a second equivalent of silane was added after the
monohydrosilylation of benzonitrile had completed, it took
only 17 h to achieve the double hydrosilation. The reason for the
discrepancy remains unknown.
[21] We did, however, succeed in the preparation of its pyridine (py)
analogue, [Cp(iPr3P)RuH(py)]: D. V. Gutsulyak, G. I. Nikonov,
manuscript in preparation.
[22] DFT studies show that silyl migration on the coordinated nitrile
in 2 does not occur (see Ref. [15]).
7556
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7553 –7556