A highly active manganese precatalyst for the hydrosilylation of ketones and esters

Article abstract of DOI:10.1021/ja4116346

The reduction of (Ph₂ ^PPrPDI)MnCl₂ allowed the preparation of the formally zerovalent complex, (Ph₂ ^PPrPDI)Mn, which features a pentadentate bis(imino)pyridine chelate. This complex is a highly active precatalyst for the hydrosilylation of ketones, exhibiting TOFs of up to 76,800 h^-1 in the absence of solvent. Loadings as low as 0.01 mol % were employed, and (Ph₂ ^PPrPDI)Mn was found to mediate the atom-efficient utilization of Si-H bonds to form quaternary silane products. (Ph₂^PPrPDI)Mn was also shown to catalyze the dihydrosilylation of esters following cleavage of the substrate acyl C-O bond. Electronic structure investigation of (Ph ₂^PPrPDI)Mn revealed that this complex possesses an unpaired electron on the metal center, rendering it likely that catalysis takes place following electron transfer to the incoming carbonyl substituent.

Full text of DOI:10.1021/ja4116346

Communication
pubs.acs.org/JACS
A Highly Active Manganese Precatalyst for the Hydrosilylation of
Ketones and Esters
Tufan K. Mukhopadhyay, Marco Flores, Thomas L. Groy, and Ryan J. Trovitch*
Department of Chemistry and Biochemistry, Arizona State University, Tempe, Arizona 85287, United States
S
* Supporting Information
have since been reported to mediate ketone hydrosilylation,
ABSTRACT: The reduction of (Ph PPrPDI)MnCl2 allowed
both catalysts operate with modest TOFs (up to 40 h−1).19
Recent studies have shown (3,5-tBu2-salen)MnN to be an
effective catalyst for the hydrosilylation of aldehydes and ketones
at elevated temperatures20 and that Mn2(CO)10 can be irradiated
with UV light to achieve the dihydrosilylation of carboxylic
acids.21 In this report, we describe a redox-active, ligand-
supported Mn complex that catalyzes ketone hydrosilylation at
25 °C, with TOFs that are 45× greater than those for
(Ph3P)(CO)4MnC(O)Me.16 Comparatively, the Mn complex
described herein is roughly 3× more active than the best Fe
catalyst,6a 1,500× more active than the best Co catalyst,22 and
300× more active than the best Ni catalyst23 for carbonyl
hydrosilylation under mild conditions. This catalyst has also been
found to reductively cleave the acyl C−O bond of esters during
their dihydrosilylation to yield a mixture of silyl ethers.
2
the preparation of the formally zerovalent complex,
(
Ph PPrPDI)Mn, which features a pentadentate bis(imino)-
2
pyridine chelate. This complex is a highly active precatalyst
for the hydrosilylation of ketones, exhibiting TOFs of up
to 76,800 h−1 in the absence of solvent. Loadings as low as
0.01 mol % were employed, and (Ph PPrPDI)Mn was found
2
to mediate the atom-efficient utilization of Si−H bonds to
form quaternary silane products. (Ph PPrPDI)Mn was also
2
shown to catalyze the dihydrosilylation of esters following
cleavage of the substrate acyl C−O bond. Electronic
structure investigation of (Ph PPrPDI)Mn revealed that this
2
complex possesses an unpaired electron on the metal
center, rendering it likely that catalysis takes place
following electron transfer to the incoming carbonyl
substituent.
This study began with preparing (Ph PPrPDI)MnCl2 (1, eq 1),
2
following the addition of Ph PPrPDI24 to (THF)2MnCl2 in toluene
2
solution. The complex was found to have a magnetic moment of
6.0 μB at 23 °C (SMn = 5/2), and single-crystal X-ray diffraction
revealed that 1 possesses an unreduced κ3-N,N,N-PDI chelate25
in the solid state (Figure S1 and Table S2 of Supporting
Information [SI]). Reduction of 1 using excess Na/Hg in the
presence of 1,3,5,7-cyclooctatetraene (COT) afforded the
ydrosilylation is conducted for the manufacture of silicone-
Hbased materials including coatings, adhesives, and cured
rubbers.1 Although this transformation has long relied on the use
of precious metal catalysts,2 the relative toxicity and cost of these
metals has inspired the search for sustainable alternatives that
exhibit comparable activity (or turnover frequency, TOF),
selectivity, and stability (or turnover number, TON).3 Since
traditional coordination compounds of Mn, Fe, and Co tend to
participate in one-electron reaction pathways, rather than well-
defined oxidative addition and reductive elimination cycles,4 the
development of first-row metal hydrosilylation catalysts has only
recently accelerated.5,6 The use of redox non-innocent ligands
has emerged as a leading method to counteract this
inconvenience.7,8 With the use of sterically demanding bis-
(imino)pyridine (or pyridine diimine, PDI) ligands, Fe catalysts
that mediate the efficient hydrosilylation of olefins9−12 and
ketones13,14 have been developed.
formally zerovalent complex, (Ph PPrPDI)Mn (2, eq 1). This
2
reaction was found to proceed at a slower rate in the absence of
COT (completion at 60 h). Although 2 is paramagnetic (2.2 μB at
23 °C), it was found to exhibit a 1H NMR spectrum that features
10 broadened resonances that are shifted over a 200 ppm range.
In contrast, Mn hydrosilylation catalyst design has yet to
benefit from the use of redox-active chelates.15 Complexes of the
general formula (CO)5Mn(X) (X = C(O)Ph, C(O)Me, Me, Br)
have been found to hydrosilylate acetone, albeit with limited
TOFs (5−25 h−1).16 Nonetheless, a phosphine-substituted
derivative of this catalyst class, (Ph3P)(CO)4MnC(O)Me, has
emerged as the most active Mn-based ketone hydrosilylation
catalyst to date, exhibiting TOFs of 27.2 min−1 (1,632 h−1) at
ambient temperature.16 Furthermore, this complex catalyzes the
dihydrosilylation of esters to form the corresponding ether and
siloxane products.17 Although (η3-1H-hydronaphthalene)Mn-
The molecular structure of 2 (Figure 1) was determined so
that information regarding PDI reduction could be obtained. The
geometry about Mn in 2 is distorted trigonal bipyramidal with
N(2)−Mn(1)−P(1), N(2)−Mn(1)−P(2), and N(1)−Mn(1)−
N(3) angles of 130.90(6), 123.14(6), and 157.54(9)°,
Received: November 14, 2013
Published: December 24, 2013
18
(CO)3P(OMe)3 and [(η6-naphthalene)Mn(CO)3][BF4]19
© 2013 American Chemical Society
882
dx.doi.org/10.1021/ja4116346 | J. Am. Chem. Soc. 2014, 136, 882−885

Journal of the American Chemical Society
Communication
(|Ax| = 161.2, |Ay| = 375.4, |Az| = 164.8 MHz). Both properties are
consistent with the crystallographically determined coordination
environment about the Mn in 2 and are characteristic of a low-
spin Mn(II) complex.27
With 2 in hand, the ability of this complex to catalyze the
hydrosilylation of ketones at 25 °C was evaluated. Adding 1 mol
% of 2 to an equimolar solution of cyclohexanone and PhSiH3 in
benzene-d6 afforded complete ketone reduction after 4 min
(TOF = 1,485 h−1). In addition to PhSiH(OCy)2, a significant
quantity of residual PhSiH3 was identified by 1H NMR
spectroscopy. Since efficient hydrosilylation was achieved in
the presence of PhSiH3, the reductant was systematically varied
to probe whether steric or electronic effects influence the rate of
ketone reduction (Table S4, SI). While 26% conversion was
observed after 4 min with Ph2SiH2, no reaction was observed
over the same time period for tertiary silanes such as Ph3SiH or
Et3SiH. In contrast, 28% conversion was achieved after 4 min in
the presence of (EtO)3SiH (TOF = 420 h−1), confirming that
quaternary silane products can be formed when using 2 as a
catalyst. The activity noted for (EtO)3SiH (compared with that
for Ph3SiH or Et3SiH) suggests that the hydrosilylation of
cyclohexanone with PhSiH3 results in the formation of
PhSiH(OCy)2 rather than PhSiH2(OCy) since the silyl ethers
formed during ketone reduction enhance the reactivity of
remaining Si−H bonds.
Figure 1. Molecular structure of 2 shown at 30% probability ellipsoids.
H atoms omitted for clarity. Relevant distances and angles are discussed
in the text and provided in Table S3 of the SI.
Encouraged, we investigated the scope of 2-catalyzed ketone
hydrosilylation (Table 1). By using an equimolar quantity of
PhSiH3 and 1 mol % of 2, acetophenone was fully reduced after 4
min at 25 °C to yield a 3:1 ratio of PhSiH(OCH(Me)(Ph))2 to
PhSi(OCH(Me)(Ph))3 (entry 1). Varying the phenyl group of
acetophenone to include electron-donating (entries 2,3) or
-withdrawing substituents (entries 4,5) greatly extended the time
required to complete the reaction. A similar effect was observed
for 2,2,2-trifluoroacteophenone (entry 6), but this reaction
turned an uncharacteristic blue-green color upon adding 2, and a
delay in turnover onset was noticed (no reaction at 2 h). Steric
bulk about the ketone functionality was also found to
significantly influence turnover rates (entries 8−10).
Attempts to scale up this transformation were made to
demonstrate the synthetic utility of 2. Initially, the hydro-
silylation of acetophenone with PhSiH3 was conducted using 0.1
mol % of 2 in the absence of solvent (Table 1, entry 11).
Exothermic in nature, the reaction reached completion after 4
min to yield a 5:1 mixture of PhSiH(OCH(Me)(Ph))2 and
PhSi(OCH(Me)(Ph))3 (TOF = 14,850 h−1). Although the
silane products formed in this reaction proved difficult to
separate, it was found that unhindered ketones such as
cyclohexanone or 2-hexanone could be hydrosilylated after 4
min under these conditions to selectively yield either PhSiH-
(OCy)2 or PhSiH(OCH(Me)(nBu))2, respectively. Thus, the
catalyst loading was lowered to 0.01 mol %, and the 2-catalyzed
hydrosilylation of cyclohexanone or 2-hexanone with PhSiH3
reached completion after 5 min (entries 12,13) as judged by 1H
NMR spectroscopy. Because a significant amount of heat was
generated upon catalyst addition, it is believed that the isolated
yields are lower than expected due to vaporization of the
substrate [64% for PhSiH(OCy)2, 62% for PhSiH(OCH(Me)-
(nBu))2]. Thus, the TOFs of these transformations are best
calculated from the isolated yields (entry 12: TOF = 76,800 h−1;
entry 13: TOF = 74,400 h−1), even though they represent the
lower limit of what may have been achieved.
Figure 2. Experimental (solid line) and simulated (dashed line) X-band
EPR spectra of 2 at 77 K.
respectively. The PDI chelate of 2 features significantly elongated
N(1)−C(2) and N(3)−C(8) distances of 1.354(3) and 1.355(3)
Å, along with contracted C(2)−C(3) and C(7)−C(8) distances
of 1.416(4) and 1.414(3) Å, respectively. These bond lengths,25
which are largely consistent with those reported for
26
2
(
2,6‑iPr PhPDI)Mn(THF)2, suggest that 2 contains a low-spin
Mn(II) center (SMn = 1/2) that is supported by a singlet PDI
dianion. To obtain supporting evidence for this electronic
structure determination, the X-band (9.44 GHz) electron
paramagnetic resonance (EPR) spectrum of 2 was recorded in
a toluene glass at 77 K. Although a signal consistent with the
presence of Mn(II) was observed (Figure 2), the spin state of the
metal center and its hyperfine coupling interaction could not be
determined by simple inspection of the spectrum. To obtain the
EPR parameters, the respective spin Hamiltonian was fit to the
data (Figure 2, dashed line). The spectral features observed for 2
were well-fit (σ = 1.7%, see SI) considering a low-spin 55Mn
center (S = 1/2, I = 5/2) with anisotropic g values (gx = 2.079, gy
= 2.037, gz = 2.017) and large anisotropic hyperfine couplings
Since two hydrosilylation products were identified for many of
the ketones in Table 1, efforts were made to improve the
883
dx.doi.org/10.1021/ja4116346 | J. Am. Chem. Soc. 2014, 136, 882−885

Journal of the American Chemical Society
Communication
a
a
Table 1. Hydrosilylation of Ketones Using 2
Table 2. Atom-Efficient Hydrosilylation Reactions
a
b
Conversion determined by 1H NMR spectroscopy. Reaction
conducted in 0.7 mL of benzene-d6 with ∼0.0027 mmol of 2, 0.27
c
mmol of silane, and 0.81 mmol of substrate. Reaction conducted in
0.7 mL of benzene-d6 with 0.0036 mmol of 2, 0.36 mmol of silane, and
0.72 mmol of substrate.
achieved within 42 min of adding 2 (TOF = 283 h−1). When 3
equiv of 2-hexanone was added per PhSiH3 equiv, the reaction
did not reach completion after 24 h at 25 °C (entry 4), as
PhSiH(OCH(Me)(nBu))2 reacted slowly under these con-
ditions.
The ability of 2 to catalyze the hydrosilylation of esters was
also investigated (Table 3). To our surprise, adding 1 mol % of 2
to an equimolar solution of MeOAc and PhSiH3 in benzene-d6
afforded a mixture of quaternary silanes that included
PhSi(OMe)3, PhSi(OEt)3, PhSi(OMe)2(OEt), and PhSi-
(OEt)2(OMe) (entry 1). 1H and 2H NMR spectroscopy showed
deuterium incorporation into the methylene position of each
ethoxysilane product when this reaction was repeated using
PhSiD3, confirming that the ethoxide substituents arise from
carbonyl reduction following reductive cleavage of the MeOAc
acyl C−O bond. Although the reduction of MeOAc was not
complete after 16 h at 25 °C, the dihydrosilylation of EtOAc to
form a 9:1 mixture of PhSi(OEt)3 to PhSiH(OEt)2 took only 5.5
h to reach completion (entry 2). In contrast, the dihydrosilyl-
t
ation of iPrOAc, PhOAc, or BuOAc to tertiary and quaternary
silane products required up to 10 d or heating to 80 °C to reach
appreciable conversion (entries 3−5). On the basis of the results
summarized in Table 3, it should be noted that the relative rates
of ester dihydrosilylation do not closely correlate to the acyl
C−O bond cleavage preferences reported for
a
Reactions conducted in 0.7 mL of benzene-d6 with ∼0.0033 mmol of
b
28
(
2,6‑iPr PhPDI)Fe(N2)2.
2
2, 0.33 mmol of PhSiH3, and 0.33 mmol of substrate. Aryl groups
c
1
designated as Ar. Conversion determined by H NMR spectroscopy.
In contrast to the commonly evoked Chalk−Harrod or
d
Reaction conducted at 0.1 mol % of 2 in a neat solution of PhSiH3
modified Chalk−Harrod mechanisms for metal-catalyzed hydro-
silylation,3,29 it is believed that 2 catalyzes this transformation by
way of radical transfer to the incoming carbonyl moiety. This
pathway is consistent with the electronic structure investigation
of 2, which has been found to possess a Mn-localized unpaired
spin. For the hydrosilylation of ketones, radical transfer is likely
followed by the abstraction of an H atom from silane, along with
concomitant Si−O bond formation. It is believed that radical
transfer and H atom abstraction also take place during ester
hydrosilylation, but this substrate class may undergo β-alkoxide
elimination to liberate an aldehyde that is ultimately converted to
e
and substrate. Neat reaction conducted at 0.01 mol % of 2.
selectivity and atom efficiency of this transformation by
decreasing the relative amount of PhSiH3 added to the reaction.
Adding 0.33 mol % of 2 to a benzene-d6 solution containing 3
equiv of cyclohexanone or acetophenone per equiv of PhSiH3
resulted in complete Si−H bond utilization and the selective
formation of PhSi(OCy)3 or PhSi(OCH(Me)(Ph))3, respec-
tively (Table 2, entries 1, 2). Likewise, the hydrosilylation of
diisopropyl ketone to selectively yield PhSiH(OCH(iPr)2)2 was
884
dx.doi.org/10.1021/ja4116346 | J. Am. Chem. Soc. 2014, 136, 882−885

Journal of the American Chemical Society
Communication
funded by the U.S. Department of Energy, Office of Science,
Basic Energy Sciences under Award Number DE-SC0001016.
Table 3. Reductive Cleavage and Dihydrosilylation of Esters
Using 2
a
REFERENCES
■
(1) Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis: The Applications
and Chemistry of Catalysis by Soluble Transition Metal Complexes, 2nd ed.;
John Wiley & Sons, Inc.: New York, 1992; pp 39−41.
(2) Speier, J. L.; Webster, J. A.; Barnes, G. H. J. Am. Chem. Soc. 1957,
79, 974−979.
(3) Troegel, D.; Stohrer, J. Coord. Chem. Rev. 2011, 255, 1440−1459.
(4) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Sausalito, 1987; pp 306−333.
(5) Reviews: (a) Chakraborty, S.; Guan, H. Dalton Trans. 2010, 39,
7427−7436. (b) Enthaler, S.; Junge, K.; Beller, M. Angew. Chem., Int. Ed.
2008, 47, 3317.
(6) For highly active bis(amide)iron precatalysts see: (a) Ruddy, A. J.;
Kelly, C. M.; Crawford, S. M.; Wheaton, C. A.; Sydora, O. L.; Small, B.
L.; Stradiotto, M.; Turculet, L. Organometallics 2013, 32, 5581−5588.
(b) Yang, J.; Tilley, T. D. Angew. Chem., Int. Ed. 2010, 49, 10186−10188.
(7) Lyaskovskyy, V.; de Bruin, B. ACS Catal. 2012, 2, 270−279.
(8) Chirik, P. J.; Wieghardt, K. Science 2010, 327, 794−795.
(9) Atienza, C. C. H.; Tondreau, A. M.; Weller, K. J.; Lewis, K. M.;
Cruse, R. W.; Nye, S. A.; Boyer, J. L.; Delis, J. G. P.; Chirik, P. J. ACS
Catal. 2012, 2, 2169−2172.
(10) Tondreau, A. M.; Atienza, C. C. H.; Darmon, J. M.; Milsmann, C.;
Hoyt, H. M.; Weller, K. J.; Nye, S. A.; Lewis, K. M.; Boyer, J.; Delis, J. G.
P.; Lobkovsky, E.; Chirik, P. J. Organometallics 2012, 31, 4886−4893.
(11) 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.
(12) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2004, 126,
13794−13807.
(13) Tondreau, A. M.; Darmon, J. M.; Wile, B. M.; Floyd, S. K.;
Lobkovsky, E.; Chirik, P. J. Organometallics 2009, 28, 3928−3940.
(14) Tondreau, A. M.; Lobkovsky, E.; Chirik, P. J. Org. Lett. 2008, 10,
2789−2792.
a
Reactions conducted in 0.7 mL of benzene-d6 with ∼0.003 mmol of
b
2, 0.3 mmol of silane, and 0.3 mmol of substrate. Conversion
c
determined by 1H NMR spectroscopy. Reaction conducted at 25 °C.
d
Reaction conducted at 80 °C.
(15) Díez-Gonzal
523−559.
́
ez, S.; Nolan, S. P. Org. Prep. Proced. Int. 2007, 39,
the respective silyl ether. A comprehensive study to determine
how the redox non-innocent Ph PPrPDI ligand aids 2-mediated
2
(16) Cavanaugh, M. D.; Gregg, B. T.; Cutler, A. R. Organometallics
1996, 15, 2764−2769.
ketone and ester hydrosilylation is underway.
(17) Mao, Z.; Gregg, B. T.; Cutler, A. R. J. Am. Chem. Soc. 1995, 117,
10139−10140.
In conclusion, the hydrosilylation of ketones and esters was
achieved under mild conditions using a redox-active ligand-
supported Mn catalyst. While excellent activities (TOFs up to
76,800 h−1) and high atom efficiencies have been observed for
the reduction of unhindered ketones, the dihydrosilylation of
esters was found to proceed through a reductive acyl C−O bond
cleavage pathway with relatively modest TOFs. Even though Mn-
based reduction catalysts remain underexplored and underutil-
ized, it is hoped that the transformations reported herein
highlight the potential value of high-denticity, redox non-
innocent chelates in catalyst design.
(18) Son, S. U.; Paik, S.-J.; Lee, I. S.; Lee, Y.-A.; Chung, Y. K.; Seok, W.
K.; Lee, H. N. Organometallics 1999, 18, 4114−4118.
(19) Son, S. U.; Paik, S.-J.; Chung, Y. K. J. Mol. Catal. A: Chem. 2000,
151, 87−90.
(20) Chidara, V. K.; Du, G. Organometallics 2013, 32, 5034−5037.
(21) Zheng, J.; Chevance, S.; Darcel, C.; Sortais, J.-B. Chem. Commun.
2013, 49, 10010−10012.
(22) Niu, Q.; Sun, H.; Li, X.; Klein, H.-F.; Florke, U. Organometallics
̈
2013, 32, 5235−5238.
(23) Chakraborty, S.; Krause, J. A.; Guan, H. Organometallics 2009, 28,
582−586.
(24) Ben-Daat, H.; Hall, G. B.; Groy, T. L.; Trovitch, R. J. Eur. J. Inorg.
Chem. 2013, 4430−4442.
ASSOCIATED CONTENT
■
S
* Supporting Information
(25) Knijnenburg, Q.; Gambarotta, S.; Budzelaar, P. H. M. Dalton
Trans. 2006, 5442−5448.
This material is available free of charge via the Internet at http://
pubs.acs.org.
(26) Russell, S. K.; Bowman, A. C.; Lobkovsky, E.; Wieghardt, K.;
Chirik, P. J. Eur. J. Inorg. Chem. 2012, 535−545.
(27) Rieger, P. H. Coord. Chem. Rev. 1994, 135/136, 203−286.
(28) Trovitch, R. J.; Lobkovsky, E.; Bouwkamp, M. W.; Chirik, P. J.
Organometallics 2008, 27, 6264−6278.
AUTHOR INFORMATION
■
Corresponding Author
ryan.trovitch@asu.edu
(29) Chalk, A. J.; Harrod, J. F. J. Am. Chem. Soc. 1965, 87, 16−21.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported as part of the Center for Bio-Inspired
Solar Fuel Production, an Energy Frontier Research Center
■
885
dx.doi.org/10.1021/ja4116346 | J. Am. Chem. Soc. 2014, 136, 882−885