Cobalt-catalyzed hydrosilation/hydrogen-transfer cascade reaction: A new route to silyl enol ethers

Article abstract of DOI:10.1002/chem.201301502

Capitalizing on cobalt: A new route to silyl enol ethers employing a Co-catalyzed cascade reaction featuring a tandem hydrosilation/hydrogen-transfer reaction is reported. The low catalyst loading, mild reaction conditions, and unique η²-silane resting state showcase the impressive utility of this seldom used transition-metal catalyst in C-H activation reactions (see scheme; VTMS = vinyltrimethylsilane; Cp* = 1,2,3,4,5- pentamethylcyclopentadiene). Copyright

Full text of DOI:10.1002/chem.201301502

DOI: 10.1002/chem.201301502
Cobalt-Catalyzed Hydrosilation/Hydrogen-Transfer Cascade Reaction: A
New Route to Silyl Enol Ethers
Thomas W. Lyons and Maurice Brookhart*[a]
Silyl enol ethers are useful synthetic intermediates for a
as 1) the reducing agent for the ketone; 2) a directing group
for the Co catalyst; 3) as the hydrogen acceptor for hydro-
gen transfer. In initial experiments, we were pleased to dis-
[1]
À
À
variety of C C and C X bond-forming processes. Tradi-
tional approaches to silyl enolate formation involve O-silyla-
tion of lithium enolates.[2,3] These methods are incompatible
with certain acidic functional groups and can require the use
of protecting groups to prevent deleterious reactivity. Addi-
tionally, stoichiometric amounts of by-products and salts are
formed as a result of these approaches. Recent transition-
metal-catalyzed methods to overcome the requirement for
strong bases have included Ir-catalyzed carbonylative silyla-
tion,[4] Rh-catalyzed aldehyde homologation,[5] and Cu-cata-
lyzed silyl migration.[6] Dehydrogenative silylation of ke-
tones has also been achieved by using Rh[7] and Ru[8] cata-
lysts.
cover [CoCp*ACTHUNRGTNEUNG(VTMS)2] catalyzed the desired reaction at
room temperature overnight. By using one equivalent of
silane, the reaction did not achieve completion after 8 h
(Table 1, entry 1). Increased loading of silane (R=Me) re-
sulted in full conversion (Table 1, entries 2 and 3). The reac-
Table 1. Optimization of silyl enol ether formation.[a]
Recent work from our lab has shown that the [CoCp*-
Entry
G
R
Silane
[equiv]
Solvent
t
Conversion[b]
[%]
ACHTUNGTRENNUNG(VTMS)2] complex 1 (VTMS=vinyltrimethylsilane) can be
3
A
[h]
À
used in the activation of sp C H bonds for the synthesis of
enamines.[9] Similar complexes have been used in hydrosila-
tion[10] and in hydrogen-transfer reactions.[11] We postulated
that with the appropriate silane reagent, 1 could catalyze a
tandem hydrosilation/hydrogen-transfer cascade to produce
silyl enol ethers [Eq. (1)]. This approach is appealing for
two reasons. First, the reaction could be carried out under
base-free conditions, opening the possibility of new substrate
scopes. Second, the reaction would be atom economical,
producing no stoichiometric salts or by-products. Herein, we
report successful development of this methodology for the
synthesis of silyl enol ethers. Additionally, we provide NMR
evidence to support a unique h2-vinyl silane complex as the
catalyst resting state during this transformation.
1
2
3
4
5
6
7
8
9
Me
Me
Me
Me
Me
Me
Me
Me
Ph
1
toluene
toluene
toluene
benzene
CDCl3
CD2Cl2
toluene
toluene
toluene
toluene
8
8
8
71
95
100
1.5
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.5
8
100
15
15
9
18
9
decomp
decomp
99
82
100
99
10
Ph
15
[a] Reaction
G
acetophenone
(0.076 mmol)
with
1
ated solvent (0.7 mL). [b] Monitored by H NMR spectroscopy.
tion worked equally well in benzene and toluene (Table 1,
entries 3 and 4). Halogenated solvents appeared to cause
catalyst decomposition. In these reactions the solution
1
turned green, and H NMR analysis failed to show any rec-
ognizable resonances, suggesting a paramagnetic Co com-
plex (Table 1, entries 5 and 6). The catalyst loading could be
reduced to as low as 1 mol% while still obtaining acceptable
conversions (Table 1, entries 7 and 8). Finally, the phenyl-
substituted vinyl silane (R=Ph) worked equally well, albeit
with a reduced rate (Table 1, entries 9 and 10).
With these results in hand, other substrates were explored
for this Co-catalyzed hydrosilation/hydrogen-transfer reac-
tion by using both Me2ACTHUNTGRNENUG(vinyl)silane and MePhACHTUNGTERN(NUGN vinyl)silane.
Initial investigations into this approach focused on using
acetophenone substrates with commercially available vinyl
silanes. The vinyl silane serves three important roles. It acts
[a] Dr. T. W. Lyons, Prof. M. Brookhart
Department of Chemistry
University of North Carolina at Chapel Hill
Chapel Hill, NC 27599-3290 (USA)
Fax : (+1)919-962-3290
A number of alkyl-substituted acetophenones were viable
substrates including 2- and 4-methyl acetophenone, as well
as the more sterically hindered 2’,4’,6’-trimethylacetophe-
E-mail: mbrookhart@unc.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201301502.
10124
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10124 – 10127

COMMUNICATION
Table 2. Optimization of silyl enol ether formation.[a]
the reaction (>99% conversion), a small amount of CHCl3
was added to the solution. Stirring for 5–10 min, followed by
filtration over Celite and removal of volatiles (including
excess silane), resulted in analytically pure silyl enol ethers.
Importantly, the solvents used in the workup must be kept
dry and acid free (CHCl3 stored over sieves and CaCO3) to
prevent hydrolysis of the products.
Entry
Product
R
t
Conversion
[%][b]
Yield
[%][c]
[h]
1
2
Me
Ph
9
15
>99
>99
95
99
Over the course of the catalytic reaction when using Me2-
(vinyl)SiH, a hydride signal (À21.2 ppm, broad) was ob-
served, indicating a cobalt hydride complex as a likely rest-
ing state in the catalytic cycle. To gain some insight into the
mechanism of this reaction and obtain additional informa-
tion about the unique reactivity of cobalt in hydrogen-trans-
fer reactions, stoichiometric studies were performed. The
3
4
Me
Ph
10
15
>99
>99
94
93
[CoCp*ACTHNURTGENNG(U VTMS)2] complex 1 was combined with different
amounts of silane, and the reactions were monitored by
5
6
Me
Ph
11
15
>99
>99
98
96
1H NMR spectroscopy. Addition of 1–5 equiv of RMe-
AHCTUNGTRENG(NNU vinyl)SiH (R=Me, Ph) to 1 in C6D6 revealed that only one
equivalent of silane was required to displace two equivalents
of VTMS and generate a new cobalt hydride complex
[Eq. (2)].
7
8
Me
Ph
12
14
>99
>99
95
98
9
10
Me
Ph
9
9
>99
>99
94
95
11
12
Me
Ph
10
12
>99
>99
97
98
The resulting species contained a hydride resonance con-
sistent with that observed during catalysis (d=
À21.2 ppm).[12] This new complex also contained a pair of
1H doublets at 1.58 and 1.59 ppm with coupling constants of
12.0 and 9.6 Hz, respectively, as well as a 1H ddd at 2.31
(J=12.0, 9.6, 3.6 Hz). These new signals and coupling con-
stants are consistent with a Co-coordinated vinyl moiety.
Two new diastereotopic methyl signals at d=0.62 and
À0.29 ppm (both doublets with J=1.8 Hz) implied the
silane was coordinated to cobalt as shown in 2. Although
the shifts were consistent with the complex proposed, the
splitting of the multiplet at d=2.31 ppm and two methyl
doublets indicated another proton correlation. A two-di-
13
14
Me
Ph
5
6
>99
>99
93
97
15
16
Me
Ph
12
14
>99
>99
95
99
[a] Reaction conditions: ketone (0.15 mmol) with [CoCp*ACTHNUTRGNEUNG(VTMS)2]
(3.8 mmol) and silane (0.3 mmol) in deuterated solvent (0.7 mL).
[b] Monitored by 1H NMR spectroscopy. [c] Yields of the isolated prod-
uct.
1
mensional H COSY NMR experiment revealed a weak cor-
relation between H4 and H1. To confirm this correlation, a
selective homonuclear decoupling experiment was per-
formed. With H4 decoupled, the ddd at d=2.31 ppm was re-
duced to a dd with J=12.0 and 9.6 Hz (Figure S1 in the Sup-
porting Information). Additionally, the two methyl signals at
d=0.62 and À0.29 ppm were reduced to singlets (Figure S2
and S3 in the Supporting Information). These correlations
not only explain the observed multiplet at d=2.31 ppm but
none (Table 2, entries 3–8). Halide-substituted acetophe-
nones worked equally as well in this reaction; aryl Cl and
aryl Br bonds were unreactive under the reaction condi-
tions (Table 2, entries 9–12).
À
À
Additionally,
highly
electron-deficient
compounds
(Table 2, entries 13 and 14) and polycyclic aryl ketones were
efficient substrates for silyl enol ether formation (Table 2,
entries 15 and 16).
One important advantage of this protocol is the ease and
practicality of work-up and isolation. Upon completion of
the magnitude of the coupling is also suggestive of a re-
2
À
tained Si H bond, consistent with a h -silane complex 3.
Distinguishing between the classical silyl hydride structure
2 and the h2-silane structure 3 can be accomplished by ex-
Chem. Eur. J. 2013, 19, 10124 – 10127
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10125

M. Brookhart and T. W. Lyons
1
amining the 29Si–1H coupling constants. The H NMR spec-
proton shifts of the methyl peaks in each diastereomer were
reduced to singlets in the decoupling experiment (see Fig-
À
trum of the Co H singlet at d=À21.3 ppm of
2 in
[D8]toluene is severely broadened at 208C (18 Hz width at
ure S5 in the Supporting Information). The observed JSiÀ
H
half-height) by the quadrupolar Co nucleus (I=7/2) and
coupling constants of 66 and 67 Hz confirm that with PhMe-
29
makes detection of an Si H coupling constant problematic
VinylSiH h2-silane complexes are formed (4a and b).
À
À
(Figure 1A). However, low-temperature NMR studies re-
With the structure of the Co H resting state confirmed
the following preliminary mechanism is proposed (Figure 3).
Two equivalents of VTMS were displaced from [CoCp*-
AHCTUNGTERGNUN(N VTMS)2] upon addition of RMeVinylSiH to give 3 (or 4),
Figure 1. 1H NMR signals in the hydride region at A) 208C and
B) À508C.
vealed narrowing of the signal to 10 Hz at À508C, at which
29
À
temperature the Si H satellites are clearly visible and
allow accurate determination of JSiÀ as 65 Hz (Figure 1B).
H
The value of this coupling constant is within the range of re-
lated Co h2-silane complexes, which typically fall between
40–160 Hz, whereas JSiÀ values between terminal hydride
H
and terminal silyl substituents are much smaller.[13] This sug-
gests a three-centered, two-electron interaction shown in 3
is a more accurate representation of the complex.[14]
A similar series of experiments revealed that the analo-
gous complex was formed when 1 was combined with
Figure 3. Proposed catalytic cycle for silyl enol ether formation.
PhMeACHTUNGTRENNUNG(vinyl)SiH in [D6]benzene. Two diastereomers were
observed with hydride signals at d=À20.62 and
À20.69 ppm. Two new ddd at d=2.77 and 2.45 ppm were
also observed corresponding to H1 in diastereomers 4a and
4b (Figure 2). Selective decoupling of the hydride signals
(H4) reduced the dddꢁs to ddꢁs at d=2.77 and 2.45 ppm (Fig-
ure S4 in the Supporting Information) illustrating the cou-
pling between H4 and H1 in each complex. Additionally, the
which are the resting states of the catalyst. Coordination of
À
the ketone followed by hydrosilation and O Si bond forma-
tion gave the CoI intermediate 6. The subsequent hydrogen
transfer steps most likely follow a mechanism analogous to
those proposed for related transformations.[11] C H activa-
À
tion followed by olefin insertion gives the cobaltacycle 7. b-
À
Hydride elimination and C H reductive elimination gener-
ates the silyl enol ether and regenerates the CoI catalyst.
In summary, we have disclosed a new Co-catalyzed ap-
proach for preparing silyl enol ethers from ketones. This re-
action provides a base-free and atom-economic method for
access to an important and versatile series of intermediates.
It features a new tandem hydrosilation/hydrogen-transfer
cascade with a unique h2-silane complex, characterized in
situ by NMR spectroscopy, as the catalyst resting state.
Figure 2. Proposed cobalt-hydride structures.
10126
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 10124 – 10127

Cobalt-Catalyzed Hydrosilation
COMMUNICATION
[6] a) H. Taguchi, K. Ghoroku, M. Takadi, A. Tsubouchi, T. Takeda,
Org. Lett. 2001, 3, 3811–3814; b) A. Tsubouchi, K. Onishi, T.
Takeda, J. Am. Chem. Soc. 2006, 128, 14268–14269.
[7] C. Thiot, A. Wagner, C. Mioskowski, Org. Lett. 2006, 8, 5939–5942.
[8] a) C. D. F. Kçnigs, H. F. T. Klare, Y. Ohki, K. Tatsumi, M. Oestreich,
Org. Lett. 2012, 14, 2842–2845; b) R. Gao, C. S. Yi, ACS Catal.
2011, 1, 544–547.
Acknowledgements
This work was supported by NSF under the CCI Center for Enabling
New Technologies through Catalysis (CENTC) Phase II Renewal, CHE-
1205189. T.W.L. acknowledges support from the NSF-American Compet-
itiveness in Chemistry postdoctoral fellowship (CHE-1136951). We thank
Chen Cheng for initial catalyst synthesis.
[9] A. D. Bolig, M. Brookhart, J. Am. Chem. Soc. 2007, 129, 14544–
14545.
[10] M. Brookhart, B. E. Grant, J. Am. Chem. Soc. 1993, 115, 2151–2156.
[11] M. M. Diaz-Requejo, D. DiSalvo, M. Brookhart, J. Am. Chem. Soc.
2003, 125, 2038–2039.
À
Keywords: cascade reactions · C H activation · domino
reactions · heterogeneous catalysis · hydrosilation · silyl enol
ether
[12] Over the course of these experi-
ments, a different complex was
observed in varying amounts.
2D 1H COSY NMR and inde-
pendent synthesis revealed this
[1] a) P. Brownbridge, Synthesis 1983, 85–104; b) B. A. Trofimov, L. N.
Sobenina, A. P. Demenev, A. I. Mikhaleva, Chem. Rev. 2004, 104,
2481–2506; c) S. Cacchi, G. Fabrizi, Chem. Rev. 2005, 105, 2873–
2920.
compound to be 5. Purification of
the silane by distillation under
rigorously dry conditions elimi-
nates the formation of this cata-
[2] a) P. Brownbridge, Synthesis 1983, 1–28; b) T. H. Chan in Compre-
hensive Organic Synthesis, Vol. 2 (Eds.: B. M. Trost, I. Fleming), Per-
gamon Press, Oxford, 1991, p.595.
[3] a) G. Stork, P. F. Hudrlik, J. Am. Chem. Soc. 1968, 90, 4462–4464;
b) H. O. House, L. J. Czuba, M. Gall, H. D. Olmstead, J. Org. Chem.
1969, 34, 2324–2336.
[4] a) N. Chatani, S. Ikeda, K. Ohe, S. Murai, J. Am. Chem. Soc. 1992,
114, 9710–9711; b) I. Kownacki, B. Marciniec, K. Szubert, M. Ku-
bicki, Organometallics 2005, 24, 6179–6183.
[5] a) V. K. Aggarwal, C. G. Sheldon, G. J. Macdonald, W. P. Martin, J.
Am. Chem. Soc. 2002, 124, 10300–10301; b) E. L. Dias, M. Brook-
lytically inactive complex.
[13] a) M. D. Doherty, B. Grant, P. S. White, M. Brookhart, Organome-
tallics 2007, 26, 5950–5960; b) S. C. Bart, E. Lobkovsky, P. J. Chirik,
J. Am. Chem. Soc. 2004, 126, 13794–13807; c) E. Colomer, R. J. P.
Corriu, C. Marzin, A. Vioux, Inorg. Chem. 1982, 21, 368–373;
d) J. Y. Corey, J. Braddock-Wilking, Chem. Rev. 1999, 99, 175–292;
e) J. Y. Corey, Chem. Rev. 2011, 111, 863–1071.
[14] M. Brookhart, B. E. Grant, C. P. Lenges, M. H. Prosenc, P. S. White,
Angew. Chem. 2000, 112, 1742–1745; Angew. Chem. Int. Ed. 2000,
39, 1676–1679.
Received: April 19, 2013
A
Published online: June 18, 2013
Chem. Eur. J. 2013, 19, 10124 – 10127
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10127