A Highly Chemoselective Cobalt Catalyst for the Hydrosilylation of Alkenes using Tertiary Silanes and Hydrosiloxanes

Article abstract of DOI:10.1021/acscatal.6b01091

The hydrosilylation of alkene substrates bearing additional functionalities is difficult to achieve using earth-abundant catalysts and has not been extensively realized with both earth-abundant transition metals and tertiary silanes or hydrosiloxanes. Reported herein is a well-defined bis(carbene) cobalt(I)-dinitrogen complex for the efficient, catalytic anti-Markovnikov hydrosilylation of terminal alkenes, featuring a broad substrate scope. Alkenes containing hydroxyl, amino, ester, epoxide, ketone, formyl, and nitrile groups are selectively hydrosilylated in this reaction sequence. Multinuclear NMR studies of reactive intermediates gave insights into the mechanism.

Full text of DOI:10.1021/acscatal.6b01091

Letter
pubs.acs.org/acscatalysis
A Highly Chemoselective Cobalt Catalyst for the Hydrosilylation of
Alkenes using Tertiary Silanes and Hydrosiloxanes
Abdulrahman D. Ibrahim, Steven W. Entsminger, Lingyang Zhu, and Alison R. Fout*
School of Chemical Sciences, University of Illinois at Urbana−Champaign, 600 South Mathews Avenue, Urbana, Illinois 61801,
United States
S
* Supporting Information
ABSTRACT: The hydrosilylation of alkene substrates bearing
additional functionalities is difficult to achieve using earth-
abundant catalysts and has not been extensively realized with
both earth-abundant transition metals and tertiary silanes or
hydrosiloxanes. Reported herein is a well-defined bis(carbene)
cobalt(I)-dinitrogen complex for the efficient, catalytic anti-
Markovnikov hydrosilylation of terminal alkenes, featuring a
broad substrate scope. Alkenes containing hydroxyl, amino,
ester, epoxide, ketone, formyl, and nitrile groups are selectively
hydrosilylated in this reaction sequence. Multinuclear NMR
studies of reactive intermediates gave insights into the
mechanism.
KEYWORDS: cobalt, hydrosilylation, alkenes, tertiary silanes, hydrosiloxanes, chemoselective
he catalytic hydrosilylation of alkenes is a formidable tool
alities.15 However, hydrosilylation with tertiary silanes
Tfor the synthesis of organosilicon reagents, a class of proceeded in low yields (∼30%) and substrates featuring
hydroxyl groups were not tolerated.15 The report by Nagashima
compounds widely employed in the production of consumer
features the use of tertiary hydrosiloxanes5 with an ill-defined
cobalt or iron catalyst and Chirik reports an expanded substrate
scope with a cobalt system,16 but the functional group tolerance
of both examples could be broadened.
goods and commodity chemicals.1−3 Nevertheless, the high
cost, irretrievability, and, at times, orthogonal reactivity of
precious metal catalysts have inspired efforts to develop
alternative platforms that feature a broad substrate scope and
earth-abundant metals such as cobalt, iron, or nickel.4 In
addition to the metal center, the silane featured in the catalysis
is of great importanceas tertiary silanes are more difficult to
use and hydrosiloxanes are of particular interest industrially.5
Hydrosilylation featuring earth-abundant catalysts has garnered
a significant amount of attention from several research groups,
but improvements to functional group tolerance with tertiary
silanes within these systems is still a formidable endeavor.
Earth-abundant hydrosilylation catalysts featuring iron have
been investigated by Chirik,6,7 Huang,8 Thomas,9 Ritter,10 and
others.5 Likewise, nickel-catalyzed hydrosilylation has experi-
enced a resurgence of interest.11−13 Recent reports from the
laboratories of Holland,14 Hu,15 Nagashima,5 and Chirik16 have
provided powerful new catalysts that address some of the
known hydrosilylation limitations. The CoI β-diketiminate
complexes reported by Holland and co-workers were shown to
facilitate the chemoselective hydrosilylation of alkenes in the
presence of a variety of functional groups.14 However, the
hydrosilylation of substrates featuring ketone or protic
functionalities, such as alcohols or NHx groups, was absent.14
Similarly, Hu reports a highly chemoselective [(MeN2N)-
NiOMe] pincer complex for the hydrosilylation of alkenes in
the presence of traditionally challenging functionalities such as
esters, ketones, primary amines, and even formyl function-
Because we have been interested in exploring the
competency of low-valent cobalt complexes, such as our
recently reported CoI dinitrogen compound, (DIPPCCC)CoN2
(1), (DIPPCCC = bis(diisopropylphenyl-imidazol-2-ylidene)-
phenyl),17 toward catalysis, our initial endeavors focused on
the hydrosilylation of alkenes. Our addition into this well-
established research area5,14,18−20 is to improve functional
group tolerance with tertiary silanes and hydrosiloxanes.
Herein, we report a highly chemoselective CoI catalyst for
the hydrosilylation of terminal alkenes that is amenable to the
use of both secondary and tertiary silanes in high yields. In
addition to substrates containing ketone, ester, and amine
functionalities, we demonstrate the selective 1,2-hydrosilylation
of alkenes in the presence of aldehyde, hydroxyl, and nitrile
groups, as well as conjugated dienes.
Initial hydrosilylation experiments utilized 1-octene as a
model substrate to optimize catalyst loading and solvent choice
(Table S1). In all cases, the anti-Markovnikov product was
exclusively observed at room temperature. An addition of a
drop of Hg to the reaction did not inhibit reactivity, suggesting
Received: April 17, 2016
Revised: April 29, 2016
© XXXX American Chemical Society
3589
DOI: 10.1021/acscatal.6b01091
ACS Catal. 2016, 6, 3589−3593

ACS Catalysis
Letter
that catalysis is likely homogeneous and not supported by
colloids or nanoparticles of cobalt metal. As depicted in Table
1, various silanes were compared at room temperature under
optimized conditions (5 mol % catalyst loading in benzene).
The reaction was amenable to the use of secondary (Entry 2)
and tertiary silanes, with low conversion for Et3SiH (Entry 3)
but good yields with Me2PhSiH (Entry 4) and hydrosiloxanes
such as MD′M (Entry 5) (MD′M = 1,1,1,3,5,5,5-heptamethyl-
trisiloxane) and trimethoxysilane (Entry 6). Running the
reactions in neat silane did result in the desired hydrosilylated
product but was impeded by catalyst solubility. All reactions
were run in duplicate and yields were analyzed by gas
chromatography and mass spectrometry.
Table 1. Hydrosilylation with 1-Octene
entry
[Si]
solvent
time
GC-yield
Encouraged by these results, we sought to extend the
substrate scope of the catalytic reaction sequence using both
Me2PhSiH and MD′M (Table 2). Terminal alkenes bearing a
wide variety of functional groups were investigated. In the
presence of more than one olefinic bond, the less-hindered
double bond is selectively hydrosilylated (2a). Unfortunately,
cyclic alkenes, such as cyclohexene (2b), and substrates such as
1
2
3
4
5
6
Ph3SiH
Ph2SiH2
benzene
benzene
benzene
benzene
benzene
benzene
1 h
1 h
21 h
1 h
1 h
1 h
<5%
88%
Et3SiH
<35%
82%
PhMe2SiH
MD′M
(OMe)3SiH
41%
a
94%
a
NMR yield.
c
Table 2. Substrate Scope
a
b
c
Heated at 70 °C and with 2 equiv of isoprene. Solvent was 1,4 dioxane. Isolated yields are an avg. of duplicate runs. All reactions performed on
0.56 mmol scale with a 1:1 mol mixture of alkene and silane in 3 mL of benzene unless otherwise noted.
3590
DOI: 10.1021/acscatal.6b01091
ACS Catal. 2016, 6, 3589−3593

ACS Catalysis
Letter
limonene (2c), which feature a 1,1-disubstituted olefin, are not
reactive toward hydrosilylation.
Scheme 1. Hydrosilylation of Aldehydes versus Alkenes
We postulated that such steric constraints can be leveraged to
selectively hydrosilylate more challenging substrates. Given the
significant challenges associated with the 1,2-hydrosilylation of
conjugated dienes, we turned our attention to isoprene, an
industrially relevant feedstock used in the production of natural
rubber.21 We reasoned that the steric constraints afforded by
the disubstituted alkene and conferred by our sterically
encumbered catalyst would favor hydrosilylation at the less-
hindered terminal olefinic bond. In line with our hypothesis,
hydrosilylation of isoprene proceeded with 1,2-selectivity (2e).
As expected, the analogous reaction with 2.4 equiv of 1,3-
butadiene, which possesses equally accessible double bonds,
furnished the 1,4-hydrosilylation product, as detected in the 1H
NMR spectrum. These results corroborated the importance of
sterics in modulating catalytic selectivity. Selective 1,4-hydro-
silylation of dienes has been reported with a number of
catalysts, attributed to the thermodynamic favorability of π−
allyl intermediates.22 However, very few examples of selective
1,2-hydrosilylation have been reported, and 2e signifies a rare
example where steric control leads to selectivity with a well-
defined cobalt catalyst.22−24
Primary amines (2f), tertiary amines (2g), and nitriles
(2h,3h) were also tolerated. Extending this protocol to 4-
vinylaniline was successful, resulting in alkene-selective hydro-
silylation to give 4-(2-(dimethyl(phenyl)silyl)ethyl)aniline (2f)
in moderate yield (62%). Although 4-cyanostyrene was not
productive toward the formation of the desired organosilicon
product (2i), 4-pentenenitrile was amenable to hydrosilylation
with both Me2PhSiH (2h) and MD′M (3h). Such reactivity has
been observed with precious metal catalysts25,26 but has not
been reported with cobalt.
Encouraged by these results, we next turned our attention to
the hydrosilylation of alkene substrates bearing carbonyl
functionalities. Utilizing Me2PhSiH and 5-hexen-2-one as a
model ketone-bearing substrate, the corresponding organo-
silicon product 6-(dimethyl(phenyl)silyl)hexan-2-one (2j) was
isolated in 97% yield. Extending the protocol to MD′M and
Ph2SiH2 furnished the corresponding organosilicon in 94% (3j)
and 86% isolated yield, respectively. In all cases, no hydro-
silylation of the ketone functional group was detected, and the
anti-Markovnikov product was exclusively formed (2j−k).
Esters (2l−m) and even a formyl-containing substrate, 2,2-
dimethyl-4-pentenal (2n), were tolerated with the latter also
resulting in a high yielding product with the more challenging
MD′M silane (3n). As of the publication of this manuscript,
this signifies only the first cobalt system that is tolerant of
formyl functionality.15
An investigation into the origin of this formyl group
tolerance suggests that the observed selectivity is not solely
attributable to steric considerations. Hydrosilylation of 10-
undecenal, a substrate bearing a more sterically accessible
aldehyde group, furnished two major products in a 1.4:1 ratio
favoring hydrosilylation of the aldehyde over the alkene
(Scheme 1). A competition experiment, between 1-octene
and octanal, however, initially favors hydrosilylation of the
alkene (1.6:1) but gradually converges to a 1:1 ratio over the
course of 18 h.
(phenyl)silyl)butan-1-ol in 95% yield (2o). The hydrosi-
lylation of 5-hexen-1-ol with Me2PhSiH (2p) and MD′M
(3p) proceeded in similar fashion in 1,4-dioxane. Although
tolerated with certain Pt and Rh catalysts,27,28 to the best of our
knowledge, alcohols have not been previously reported with
cobalt hydrosilylation catalysts. Further elaboration of the
catalyst’s ability to tolerate oxygen-containing substrates led us
to investigate the hydrosilylation of substrates bearing methoxy,
allyl ether, and epoxide moieties (2q−s). Gratifyingly, hydro-
silylation of these substrates proceeded in excellent yields.
Undesirable isomerization products or cleavage of the epoxide
were not observed.
Interested in gleaning insights into the mechanism of
hydrosilylation, we next investigated the stoichiometric
reactivity of the (DIPPCCC)CoN2 catalyst toward silanes. We
hypothesized that treatment of the CoI complex with silane
would result in oxidative addition of a Si−H bond to the metal
center, similar to what has been observed in recent reports by
Arnold29 and Chirik30 with low-valent, electron-rich cobalt
complexes. Such a species could then proceed along a Chalk−
Harrod reaction profile to hydrosilylate unsaturated carbon−
carbon bonds.31 Monitoring of the reaction of PhSiH3 with
(
DIPPCCC)CoN2 by 1H NMR spectroscopy (Figure S1)
suggested that such a hypothesis was plausible. The presence
of a broad upfield resonance at −5.76 ppm and decreased
symmetry in the 1H NMR spectrum, particularly in the
resonances attributed to the iPr groups of the ligand framework,
were consistent with the possibility of an oxidative addition of a
Si−H bond onto the metal center. However, rapid thermal
decomposition of this species to a new compound precluded
further characterization and therefore the possibility of an η2-
bound Si−H bond could not be ruled out. The observation of
intermediates following addition of Me2PhSiH was similarly
limited, as no product formation was observed on the 1H NMR
time scale in both the stoichiometric addition and in the
presence of excess silane.
Interestingly, addition of Ph2SiH2 to the catalyst resulted in
the formation of a more thermally robust compound amenable
to characterization. Analogous to the reaction with PhSiH3,
addition of Ph2SiH2 to the CoI catalyst rapidly resulted in the
formation of a new diamagnetic compound, 4, with an upfield
resonance at −7.08 ppm, as well as decreased symmetry of the
1
We next explored the hydrosilylation of substrates containing
unprotected alcohols, a challenging functional group for iron
and nickel catalysts.8,15 In the presence of 5 mol % of catalyst,
3-buten-1-ol was reacted with Me2PhSiH to afford 4-(dimethyl-
ligand framework in the H NMR spectrum. Two septets at
3.42 and 2.21 ppm, attributed to the methine protons of the iPr
groups of the ligand framework, and four doublets at 1.50, 1.11,
i
0.94, and 0.63 ppm, assigned to the methyls of the Pr groups,
3591
DOI: 10.1021/acscatal.6b01091
ACS Catal. 2016, 6, 3589−3593

ACS Catalysis
Letter
Figure 1. Proposed hydrosilylation mechanism (left). The crystal structure of the off-cycle product, 5, depicted with the ellipsoids at the 50%
probability level. All solvent molecules, H atoms, and the phenyl rings of the SiHPh2 fragment have been omitted for clarity (right). Select bond
lengths and angles are listed in Table S3.
are consistent with a Cs-symmetric compound. The presence of
a downfield resonance at 5.52 ppm in the H NMR spectrum
In conclusion, we have developed a highly selective CoI
system for the hydrosilylation of alkenes. This hydrosilylation,
proposed to proceed through a Chalk−Harrod type mecha-
nism, furnishes the anti-Markovnikov product exclusively in all
observed cases and is tolerant of a wide range of challenging
functionalities, including unprotected alcohols, ketones, and
dienes. The hydrosilylation of a substrate bearing formyl
functionality and an alkene bearing nitrile functionality have
also been shown to proceed selectively at the alkene.
1
was assigned to the proton of a bound silyl ligand. Additional
1
characterization with H-coupled 29Si NMR spectroscopy was
undertaken to establish the solution-state structure of 4. The
presence of a resonance at 8.28 ppm in the 29Si NMR was
assigned to a silyl ligand bound to the cobalt center. From this
peak, a 1JSiH coupling constant of 165 Hz and a smaller, weakly
resolved 2JSiH coupling constant of ∼13 Hz were measured. The
value of the latter coupling constant is below the 20 Hz JSiH
coupling value threshold associated with η2-bound Si−H
bonds32 and is consistent with cleavage of the Si−H bond by
the cobalt metal center. This was further corroborated by a
two-dimensional 1H−29Si HMBC experiment, where the
resonance centered at 8.28 ppm in the 29Si NMR exhibits a
cross-peak with the hydride, and a one-dimensional NOE
experiment (Figures S2 and S4), allowing for the assignment of
4 as the CoIII silyl-hydride compound, (DIPPCCC)Co(SiHPh2)-
(H)(N2). Complex 4 is catalytically relevant as the addition of
1-octene furnished the hydrosilylated compound and regen-
erated the (DIPPCCC)CoN2 (1) starting material.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.6b01091.
Spectral data and synthesis for complexes 2−5 (PDF)
Selected crystallographic data for complex 5 (CIF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: fout@illinois.edu.
■
Cooling a concentrated reaction mixture of (DIPPCCC)CoN2
and Ph2SiH2 in diethyl ether to −35 °C yielded yellow crystals
suitable for single-crystal X-ray diffraction. Surprisingly, an
octahedral CoIII bis(silyl) compound, (DIPPCCC)Co-
(SiHPh2)2(N2) (5), a minor decomposition product noted in
1H NMR spectra of 4 was obtained. Subsequent attempts to
independently synthesize 5 were unsuccessful, accompanied in
all cases by the concomitant formation of 4, which prohibited
the effective characterization of the CoIII bis(silyl) species. This
off-cycle decomposition product, 5, supports cleavage of the
Si−H bond, as formation of 5 is likely the result of a σ-bond
metathesis with 4 and an η2-bound silane (C). This series of
experiments has led to the proposal of a Chalk-Harrod type
mechanism (Figure 1) whereby oxidative addition of the silane
(4) is followed by coordination of the alkene (A). Subsequent
migratory insertion of the hydride (B) or silyl ligand into the
alkene, followed by reductive elimination, furnishes the
hydrosilylated product and regenerates the CoI catalyst.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support for this work was by the NSF (13151961)
and A.R.F. is an Alfred P. Sloan Research Fellow. The authors
thank Jeffery A. Bertke and Gabriel Espinosa Martinez for
crystallographic assistance.
REFERENCES
■
(1) (a) Marciniec, B. Hydrosilylation: A Comprehensive Review on
Recent Advances; Marciniec, B., Ed.; Springer: Berlin, 2009. (b) Lewis,
L. N.; Stein, J.; Gao, Y.; Colborn, R. E.; Hutchins, G. Platinum Met.
Rev. 1997, 41, 66−75. (c) Brook, M. A. Silicon in Organic,
Organometallic and Polymer Chemistry; Wiley: New York, 2000.
(d) Marciniec, B. Coord. Chem. Rev. 2005, 249, 2374−2390.
(2) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organometallics
1983, 2, 1694−1696.
3592
DOI: 10.1021/acscatal.6b01091
ACS Catal. 2016, 6, 3589−3593

ACS Catalysis
Letter
(3) Fleming, I.; Henning, R.; Plaut, H. J. Chem. Soc., Chem. Commun.
1984, No. 1, 29−31.
(4) Chirik, P. J. In Catalysis without Precious Metals; Bullock, R. M.,
Ed.; Wiley-VCH: Weinheim, Germany, 2010; pp 83−110.
(5) Noda, D.; Tahara, A.; Sunada, Y.; Nagashima, H. J. Am. Chem.
Soc. 2016, 138, 2480−2483.
(6) Tondreau, A. M.; Atienza, C. C. H.; Weller, K. J.; Nye, S. A.;
Lewis, K. M.; Delis, J. G. P.; Chirik, P. J. Science 2012, 335, 567−570.
(7) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2004,
126, 13794−13807.
(8) Peng, D.; Zhang, Y.; Du, X.; Zhang, L.; Leng, X.; Walter, M. D.;
Huang, Z. J. Am. Chem. Soc. 2013, 135, 19154−19166.
(9) Greenhalgh, M. D.; Frank, D. J.; Thomas, S. P. Adv. Synth. Catal.
2014, 356, 584−590.
(10) Wu, J. Y.; Stanzl, B. N.; Ritter, T. J. Am. Chem. Soc. 2010, 132,
13214−13216.
(11) Kuznetsov, A.; Gevorgyan, V. Org. Lett. 2012, 14, 914−917.
(12) Steiman, T. J.; Uyeda, C. J. Am. Chem. Soc. 2015, 137, 6104−
6110.
(13) Lipschutz, M. I.; Tilley, T. D. Chem. Commun. (Cambridge, U.
K.) 2012, 48, 7146−7148.
(14) Chen, C.; Hecht, M. B.; Kavara, A.; Brennessel, W. W.;
Mercado, B. Q.; Weix, D. J.; Holland, P. L. J. Am. Chem. Soc. 2015,
137, 13244−13247.
(15) Buslov, I.; Becouse, J.; Mazza, S.; Montandon-Clerc, M.; Hu, X.
Angew. Chem. 2015, 127, 14731−14734.
(16) Schuster, C. H.; Diao, T.; Pappas, I.; Chirik, P. J. ACS Catal.
2016, 6, 2632−2636.
(17) Ibrahim, A. D.; Tokmic, K.; Brennan, M. R.; Kim, D.; Matson, E.
M.; Nilges, M. J.; Bertke, J. A.; Fout, A. R. Dalton Trans. 2016. ASAP
DOI: 10.1039/c5dt04723d.
(18) Brookhart, M.; Grant, B. E. J. Am. Chem. Soc. 1993, 115, 2151−
2156.
(19) Sun, J.; Deng, L. ACS Catal. 2016, 6, 290−300.
(20) Hilt, G.; Luers, S.; Schmidt, F. Synthesis 2004, 4, 634−638.
̈
(21) Ratcha, A.; Samart, C.; Yoosuk, B.; Sawada, H.;
Reubroycharoen, P.; Kongparakul, S. J. Appl. Polym. Sci. 2015, 132,
42688.
(22) Parker, S. E.; Borgel, J.; Ritter, T. J. Am. Chem. Soc. 2014, 136,
̈
4857−4860.
(23) (a) Swisher, J. V.; Zullig, C. J. Org. Chem. 1973, 38, 3353−3357.
(b) Ojima, I.; Kogure, T. Rev. Silicon, Germanium, Tin Lead Compd.
1981, 5, 7. (c) Ojima, I. The Hydrosilylation Reaction. In The
Chemistry of Organic Silicon Compounds; Patai, S., Rappoport, Z., Eds.;
Wiley: Chichester, 1989; Vol. 1, pp 1479−1526. (d) Andrianov, K. A.;
Bagratishvili, G. D.; Kantariya, M. L.; Sidorov, V. I.; Khananashvili, L.
M.; Tsitsishvili, G. V. J. Organomet. Chem. 1965, 4, 440−445.
(e) Nasiak, L. D.; Post, H. W. J. Organomet. Chem. 1970, 23, 91−98.
(f) Fink, W. Helv. Chim. Acta 1971, 54, 1304. (g) Rericha, R.; Capka,
M. Collect. Czech. Chem. Commun. 1974, 39, 144−153.
(24) Atienza, C. C. H.; Roy, A. K.; Chirik, P. J.; Weller, K. J.; Delis, J.
G. P.; Diao, T.: US Patent Publication US 2014/0330024 (Filed: May
6, 2014; Publication: November 6, 2014).
(25) Huckaba, A. J.; Hollis, T. K.; Reilly, S. W. Organometallics 2013,
32, 6248−6256.
(26) Sabourault, N.; Mignani, G.; Wagner, A.; Mioskowski, C. Org.
Lett. 2002, 4, 2117−2119.
́ ́
(27) Marko, I.; Sterin, S.; Buisine, O.; Berthon, G.; Michaud, G.;
Tinant, B.; Declercq, J.-P. Adv. Synth. Catal. 2004, 346, 1429−1434.
(28) Takeuchi, R.; Nitta, S.; Watanabe, D. J. Chem. Soc., Chem.
Commun. 1994, No. 15, 1777−1778.
(29) Rozenel, S. S.; Padilla, R.; Arnold, J. Inorg. Chem. 2013, 52,
11544−11550.
(30) Scheuermann, M. L.; Semproni, S. P.; Pappas, I.; Chirik, P. J.
Inorg. Chem. 2014, 53, 9463−9465.
(31) Troegel, D.; Stohrer, J. Coord. Chem. Rev. 2011, 255, 1440−
1459.
(32) Karshtedt, D.; Bell, A. T.; Tilley, T. D. Organometallics 2006, 25,
4471−4482.
3593
DOI: 10.1021/acscatal.6b01091
ACS Catal. 2016, 6, 3589−3593