Hydrosilylation of Aldehydes and Formates Using a Dimeric Manganese Precatalyst

Article abstract of DOI:10.1021/acs.organomet.7b00423

The formally zero-valent Mn dimer [(^Ph2PEtPDI)Mn]₂ has been synthesized upon reducing (^Ph2PEtPDI)MnCl₂ with excess Na/Hg. Single crystal X-ray diffraction analysis has revealed that [(^Ph2PEtPDI)Mn]₂ possesses a η⁴-PDI chelate about each Mn center, as well as η²-imine coordination across the dimer. The chelate metrical parameters suggest single electron PDI reduction and EPR spectroscopic analysis afforded a signal consistent with two weakly interacting S = ¹/₂ Mn centers. At ambient temperature in the absence of solvent, [(^Ph2PEtPDI)Mn]₂ has been found to catalyze the hydrosilylation of aldehydes at loadings as low as 0.005 mol % (0.01 mol % relative to Mn) with a maximum turnover frequency of 9,900 min^-1 (4,950 min^-1 per Mn). Moreover, the [(^Ph2PEtPDI)Mn]₂-catalyzed dihydrosilylation of formates has been found to proceed with turnover frequencies of up to 330 min^-1 (165 min^-1 relative to Mn). These metrics are comparable to those described for the leading Mn catalyst for this transformation, the propylene-bridged variant (^Ph2PPrPDI)Mn; however, [(^Ph2PEtPDI)Mn]₂ is more easily inhibited by donor functionalities. Carbonyl and carboxylate hydrosilylation is believed to proceed through a modified Ojima mechanism following dimer dissociation.

Full text of DOI:10.1021/acs.organomet.7b00423

Article
pubs.acs.org/Organometallics
Hydrosilylation of Aldehydes and Formates Using a Dimeric
Manganese Precatalyst
Tufan K. Mukhopadhyay, Chandrani Ghosh, Marco Flores, Thomas L. Groy, and Ryan J. Trovitch*
School of Molecular Sciences, Arizona State University, Tempe, Arizona 85287, United States
S
* Supporting Information
ABSTRACT: The formally zero-valent Mn dimer
[(Ph2PEtPDI)Mn]2 has been synthesized upon reducing
(
Ph2PEtPDI)MnCl2 with excess Na/Hg. Single crystal X-ray
diffraction analysis has revealed that [(Ph2PEtPDI)Mn]2
possesses a κ4-PDI chelate about each Mn center, as well as
η2-imine coordination across the dimer. The chelate metrical
parameters suggest single electron PDI reduction and EPR
spectroscopic analysis afforded a signal consistent with two
weakly interacting S = 1/2 Mn centers. At ambient temperature
in the absence of solvent, [(Ph2PEtPDI)Mn]2 has been found to
catalyze the hydrosilylation of aldehydes at loadings as low as
0.005 mol % (0.01 mol % relative to Mn) with a maximum turnover frequency of 9,900 min−1 (4,950 min−1 per Mn). Moreover,
the [(Ph2PEtPDI)Mn]2-catalyzed dihydrosilylation of formates has been found to proceed with turnover frequencies of up to 330
min−1 (165 min−1 relative to Mn). These metrics are comparable to those described for the leading Mn catalyst for this
transformation, the propylene-bridged variant (Ph2PPrPDI)Mn; however, [(Ph2PEtPDI)Mn]2 is more easily inhibited by donor
functionalities. Carbonyl and carboxylate hydrosilylation is believed to proceed through a modified Ojima mechanism following
dimer dissociation.
the reaction is heated (80 °C).17 In 2014, we found that the
INTRODUCTION
■
propylene-bridged bis(imino)pyridine (or pyridine diimine,
In laboratory and industrial settings, the reduction of carbonyl-
containing compounds remains a popular synthetic route to
organic alcohols.1 This transformation can be achieved with
inorganic hydride reagents2 or under hydrogen in the presence
of a catalyst;3 however, due to the flammable nature of these
reductants, hydrosilylation has gained traction as a mild and
operationally simple approach to CO bond reduction.4
While precious metal catalysts are known to mediate this
transformation,5 poor selectivity, toxicity, and metal cost have
prompted the search for sustainable base metal alternatives.
Despite the tendency of first-row metals to engage in one-
electron chemistry,6 many efficient Fe,7 Co,8 Ni,9 Cu,10 and
Zn11 carbonyl hydrosilylation catalysts have been reported. The
activities and lifetimes of these catalysts can rival their precious
metal counterparts; however, additional improvement of
carbonyl hydrosilylation turnover frequency (TOF) remains
difficult to achieve.
To some extent, this challenge has been met through the
development of Mn catalysts.12,13 In the 1990s, Cutler and co-
workers reported that (Ph3P)(CO)4MnC(O)Me exhibits
ketone hydrosilylation14 TOFs of up to 27 min−1 and ester
reduction15 TOFs of up to 4 min−1 at ambient temperature.
Several well-defined Mn catalysts for carbonyl hydrosilylation
have since been described;16−20 however, higher activity has
only been demonstrated in a few instances. In 2013, Chidara
and Du reported that (3,5-tBu-salen)MnN hydrosilylates
aldehydes with TOFs as high as 196 min−1, but only when
PDI) compound, (Ph2PPrPDI)Mn (Figure 1, left), mediates
Figure 1. Manganese hydrosilylation catalysts featuring κ5-
ligands.18−20
ketone hydrosilylation with TOFs of up to 76 800 h−1 (more
accurately expressed as 1280 min−1) and the dihydrosilylation
of esters (to yield silyl ethers) with modest TOFs of up to 18
h−1 at 25 °C in the absence of solvent.18 Under these
conditions, we recently reported that (Ph2PPrPDI)Mn catalyzes
aldehyde hydrosilylation with TOFs of 4900 min−1 and formate
dihydrosilylation with TOFs of 330 min−1.19 Importantly,
(
Ph2PPrPDI)Mn is believed to mediate carbonyl hydrosilylation
through a modified Ojima mechanism while a related catalyst,
Ph2PPrPDI)MnH (Figure 1, middle), achieves comparable
(
Received: June 7, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
formate hydrosilylation TOFs through a hydride insertion
mechanism.19 Efforts to develop structurally related Mn
catalysts have also led us to evaluate the hydrosilylation activity
of ethylpyridine-substituted (PyEtPDEA)Mn (Figure 1, right),
which exhibits a lower maximum benzaldehyde hydrosilylation
TOF of 2475 min−1.20 Herein, we describe our efforts to
prepare and evaluate a Mn hydrosilylation precatalyst
supported by Ph2PEtPDI,21 which possesses an ethylene-bridge
between the imine and phosphine donors.
RESULTS AND DISCUSSION
■
Heating a 1:1 mixture of Ph2PEtPDI21 and (THF)2MnCl2 in
toluene to 125 °C afforded a pale orange solid identified as
(
Ph2PEtPDI)MnCl2 (1, Scheme 1). This compound was found to
Figure 2. Molecular structure of 2 shown at 30% probability ellipsoids.
Hydrogen atoms and cocrystallized toluene molecules (4 per dimer)
are omitted for clarity. For complete list of metrical parameters, see
Table S2.
Scheme 1. Preparation of [(Ph2PEtPDI)Mn]2 (2)
N(2), and Mn(1)−N(3) distances were determined to be
2.022(4), 1.947(4), and 2.092(4) Å, respectively, and the two
Mn centers are 2.7889(14) Å apart. Considering these metrical
parameters and the observed magnetic moment, it can be
proposed that 2 features intermediate spin Mn(I) centers that
are antiferromagnetically coupled to their respective singly
reduced supporting DI moieties.
To further investigate the electronic structure of 2, a toluene
solution of this complex was prepared and analyzed by X-band
(9.40 GHz) EPR spectroscopy at 107 K (Figure 3). The
be NMR silent and exhibit an ambient temperature magnetic
susceptibility of 6.0 μB (magnetic susceptibility balance, 25 °C),
consistent with a high-spin Mn(II) environment (SMn = 5/2).
On the basis of these observations, and prior structural
18
characterization of (Ph2PPrPDI)MnCl2 and (PyEtPDI)MnCl2,20
it is reasonable to propose that 1 possesses a κ3-
N,N,N-Ph2PEtPDI chelate. Reduction of 1 using excess Na/Hg
in the presence of 1,3,5,7-cyclooctatetraene (added to facilitate
reduction)18,20 afforded a red paramagnetic complex identified
as [(Ph2PEtPDI)Mn]2 (2, Scheme 1). The 1H NMR spectrum of
2 features broadened resonances at 32.26 and 26.23 ppm
(Figure S1). A single, broad resonance was also observed in the
31P NMR spectrum at −58.49 ppm (Figure S2, likely due to
uncoordinated phosphine substituents) and the solution state
magnetic moment of this compound was found to be 3.3 μB at
25 °C, suggesting two unpaired electrons in the ground state.
To obtain structural information, a single crystal of 2 was
analyzed by X-ray diffraction. Refinement revealed a dimeric
arrangement whereby each Mn center is supported by a κ4-PDI
chelate (Figure 2). Each Mn center is also coordinated to the
imine bond of a neighboring (PDI)Mn moiety in an η2-fashion
[Mn(1)−C(8A) and Mn(1)−N(3A) are 2.233(6) and
1.977(4) Å, respectively]. This interaction results in C(8)−
N(3) elongation to 1.395(6) Å, relative to the average
uncoordinated PDI imine CN distance of 1.271(17) Å,22
indicative of significant metal-to-ligand backbonding. The
geometry about Mn is best described as distorted trigonal
bipyramidal with N(1)−Mn(1)−N(3), N(2)−Mn(1)−P(1),
and N(2)−Mn(1)−N(3A) angles of 150.85(16), 129.25(12),
and 100.30(15)°, respectively. The unbridged portion of each
PDI chelate features an elongated N(1)−C(2) distance of
1.338(6) Å and a contracted C(2)−C(3) distance of 1.425(7)
Å. Although 2 is formally zero-valent, these distances are
consistent with single electron α-diimine (DI) reduction,23
whereby the PDI-based electron density resides on the
unbridged backbone atoms. The Mn(1)−N(1), Mn(1)−
Figure 3. Experimental (solid line) and simulated (dashed line) X-
band EPR spectra of 2 in toluene at 107 K. These spectra showed
small discrepancies for the magnetic field resonances above 390 mT.
Such discrepancies, as well the differences in line intensities, might be
due to inhomogeneities present in the frozen solution (powder)
sample.
observed spectral features are consistent with the presence of
two manganese centers, i.e., a broad signal showing a multiline
pattern due to hyperfine coupling (hfc) interactions between
the magnetic moment of an unpaired electron system and the
5
magnetic moment of a neighboring 55Mn (I = /2) nuclei. On
the basis of the electronic structure model proposed above,
each Mn center was assumed to carry one net unpaired electron
that results in an electronic configuration with two unpaired
B
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
electrons present in 2. Thus, the best fit of the EPR spectrum,
corresponding to 2, was obtained considering a triplet state (S
= 1) as the ground state of the electron spin system (see the
Experimental Section for the definition of the spin Hamil-
tonian). The parameters that were obtained from the fit are
summarized in Table 1. The g values are anisotropic and reflect
Precatalyst 2 has been found to tolerate fluoro, chloro, and
bromo functionalities (entries b−d), but not the iodo
functionality of 4-iodobenzaldehyde (entry e). This substrate
did not participate in an exothermic reaction with PhSiH3 and 2
due to catalyst decomposition. Complex 2 was tolerant of the
nitro and nitrile groups of 2-nitrobenzaldehyde (entry f) and 4-
cyanobenzaldehyde (entry g), respectively, as these function-
alities were not reduced under the reaction conditions. Electron
donating substituents on the aryl ring also do not influence
efficiency, as seen in the case of p-anisaldehyde and p-
tolualdehyde (entries h−i). Heteroaromatic aldehydes were
hydrosilylated without difficulty (entries k−l), however, it
should be noted that pyridine-3-carboxaldehyde did not reach
complete conversion (88%) after 2 min. Moreover, the
aldehyde functionality of 3-cyclohexene-1-carboxaldehyde
(entry m) and citral (entry n) was not efficiently reduced
during the 2 min time frame of the experiment. Under the same
conditions, (Ph2PPrPDI)Mn was found to fully reduce the
aldehyde functionality of these substrates,19 suggesting that 2 is
more susceptible to inhibition due to olefin coordination.
Olefin hydrosilylation was not observed and attempts to
hydrosilylate 1-hexene using 2 and PhSiH3 were unsuccessful,
even at 120 °C. The neat hydrosilylation of acetophenone and
cyclohexanone using 0.05 mol % 2 (0.1 mol % relative to Mn)
was also performed and both substrates were fully converted to
a mixture of silyl ethers after 4 min (TOF = 495 and 248 min−1
relative to Mn). In general, we have found that it takes several
seconds longer for neat hydrosilylation reactions featuring 2 to
exotherm, relative to when (Ph2PPrPDI)Mn is employed as the
catalyst.
Efforts were also made to optimize 2-mediated aldehyde
hydrosilylation using a stoichiometric quantity of PhSiH3.
Lowering the catalyst loading to 0.005 mol % (0.01 mol %
relative to Mn) did not diminish benzaldehyde hydrosilylation
conversion after 2 min under neat conditions, equating to a
maximum TOF of 9,900 min−1 (4,950 min−1 per Mn). On the
basis of per Mn turnover, this activity matches what was
observed for (Ph2PPrPDI)Mn-catalyzed benzaldehyde hydro-
silylation;19 however, the same conversion rate could not be
achieved using other substrates in Table 2. For example,
repeating this procedure with 4-fluorobenzaldehyde resulted in
18% conversion after 2 min, equating to a TOF of 1800 min−1
(900 min−1 per Mn).
Table 1. Parameters Used to Fit the EPR Spectrum of 2 at
9.40 GHz and T = 107 K
parametera
2
gx
2.037
1.980
1.893
gy
gz
|D|
77.7 × 10−4 cm−1
2.0 × 10−4 cm−1
114.1 × 10−4 cm−1
43.7 × 10−4 cm−1
94.7 × 10−4 cm−1
|E|
|AxMn
|
|
i
|AyMn
|AzMn
i
i
|
a
See the Experimental Section for the definition of the fitting
parameters.
a large electron spin delocalization which is consistent with the
crystallographically determined molecular structure of 2. The
zero-field splitting (ZFS) parameters are relatively small and
show a nearly axial ZFS (i.e., D ≪ gβeB0/h, and E/D ≈ 0). This
finding indicates weak electron−electron repulsion between the
two unpaired electrons present in 2.
Compound 2 was then screened for carbonyl hydrosilylation
activity. When a neat mixture of benzaldehyde and PhSiH3 was
added to 0.05 mol % 2 (0.1 mol % relative to Mn) at room
temperature, an exothermic reaction ensued, resulting in
complete conversion to a mixture of silyl ethers after 2 min.
Hydrolysis with 10% aqueous NaOH solution followed by
extraction afforded benzyl alcohol in 93% yield (Table 2, entry
a). Thirteen additional aldehydes were screened, and the
reaction outcomes are summarized in Table 2 (entries b−n).
Table 2. Hydrosilylation of Aldehydes Using 2
Attempts to hydrosilylate benzaldehyde using AIBN (radical
initiator), Mn powder, (THF)2MnCl2, or (Ph2PEtPDI)MnCl2 as
the catalyst did not result in substrate conversion. Moreover,
performing this reaction in the presence of excess Hg0 did not
adversely affect conversion, suggesting that 2 remains
homogeneous throughout catalysis. Additionally, benzaldehyde
hydrosilylation was not inhibited when the reaction was
conducted in the dark, suggesting that phosphine dissociation
via photolysis is not required for catalysis to occur. This
contrasts recent work showing that CpMn carbonyl catalysts
achieve Si−H oxidative addition following photolytic loss of
two CO ligands.16e Adding 2 equiv of NaEt3BH to 1 did not
allow for the observation or isolation of a catalytically relevant
hydride complex.
Compound 2 was also employed as a formate dihydrosily-
lation precatalyst. When a neat equimolar mixture of methyl
formate or ethyl formate and PhSiH3 was added to 0.01 mol %
2 (0.02 mol % relative to Mn), an exothermic reaction was
observed along with >99% conversion to silyl ethers within 30
min (Table 3, entries a-b). Alkaline hydrolysis of the
C
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
yield a silyl alkoxide intermediate that undergoes reductive
elimination to afford the respective silyl ether. In the case of
carboxylates, the silyl alkoxide intermediate is believed to
undergo fast β-alkoxide elimination to yield the silyl ether
product and an equivalent of formaldehyde, as described for
Table 3. Dihydrosilylation of Formates Using 2
(
Ph2PPrPDI)Mn.19 Formaldehyde is quickly hydrosilylated under
the reaction conditions to yield methoxysilane equivalents.
Given our observations of delayed catalytic onset and slower
ethyl acetate dihydrosilylation, it is believed that dimer 2 acts as
the catalyst resting state. Our inability to observe a Mn hydride
complex upon adding of 2 equiv of NaEt3BH to 1 also suggests
that catalysis is unlikely to proceed through a straightforward
insertion mechanism analogous to the one described for
(
Ph2PPrPDI)MnH.19
corresponding products was performed; however, the low
boiling points of MeOH and EtOH did not allow for their
isolation through evaporation. Furthermore, the reaction scale
was not large enough to separate these products by short path
distillation. To address this issue, our investigation shifted to
the dihydrosilylation of higher molecular weight formates. As
shown in Table 3, 2 was found to catalyze the dihydosilylation
of octyl (entry c), isoamyl (entry d), benzyl (entry e), and p-
anisyl formate (entry f) under the same conditions. Following
workup with 10% aqueous NaOH, extraction with Et2O, and
removal of the solvent and generated MeOH, the correspond-
ing alcohols were isolated in good yield. The observed 2-
mediated formate dihydrosilylation TOFs of 330 min−1 (165
min−1 relative to Mn) are slower than those observed for
CONCLUSION
■
We have described the synthesis, electronic structure, and
hydrosilylation activity of the (PDI)Mn dimer, [(Ph2PEtPDI)-
Mn]2. Although the ethylene-bridged substituents of Ph2PEtPDI
have proven too rigid to allow for κ5-PDI coordination to Mn
following (Ph2PEtPDI)MnCl2 reduction, dimerization via η2-
imine coordination occurs to fill the coordination sphere of Mn
and stabilize formally zero-valent [(Ph2PEtPDI)Mn]2. Dimer
formation was not found to prevent carbonyl hydrosilylation or
carboxylate dihydrosilylation and competitive turnover fre-
quencies were noted for each transformation. However, it
should be noted that slower aldehyde reduction in the presence
of pyridine or olefin functionalities was observed along with
delayed onset relative to propylene-bridged (Ph2PPrPDI)Mn. On
the basis of experimental observations and our inability to
observe a Ph2PEtPDI-supported Mn hydride complex, it is
proposed that [(Ph2PEtPDI)Mn]2 dissociates into monomeric
units that mediate carbonyl and carboxylate hydrosilylation
through a modified Ojima mechanism.
(
Ph2PPrPDI)Mn on a per Mn basis19 but greater than all other
reported transition metal examples.24 The dihydrosilylation of
ethyl acetate using 1.0 mol % 2 and equimolar PhSiH3 in
benzene-d6 was also performed and reduction to PhSi(OEt)3
was observed after 7.2 h at 25 °C (TOF = 14 h−1). This is again
slower than the 5.5 h needed to complete ethyl acetate
dihydrosilylation using (Ph2PPrPDI)Mn.18
Considering our recent mechanistic investigation of
(
Ph2PPrPDI)Mn-mediated hydrosilylation19 and the observations
EXPERIMENTAL SECTION
■
described herein, it is proposed that 2 mediates carbonyl
hydrosilylation and carboxylate dihydrosilylation through a
modified Ojima mechanism (Figure 4). Upon dissociating into
monomeric (Ph2PEtPDI)Mn, Si−H oxidative addition would
afford the silyl hydride intermediate shown at the top right.
Alternatively, unproductive donor association may occur at this
point and inhibit hydrosilylation. Once silane activation takes
place, substrate coordination and insertion into Mn−H would
General Considerations. Unless otherwise stated, all synthetic
reactions were performed in an MBraun glovebox under an
atmosphere of purified nitrogen. Aldrich anhydrous solvents were
purified using a Pure Process Technology solvent system and stored in
the glovebox over activated 4 Å molecular sieves and potassium
(purchased from Sigma-Aldrich) before use. Benzene-d6 was
purchased from Cambridge Isotope Laboratories and dried over 4 Å
molecular sieves before use. 1,3,5,7-Cyclooctatetraene and 2-
(diphenylphosphino)ethylamine were purchased from Strem Chem-
icals Inc. and used as received. Manganese powder, metallic mercury,
benzaldehyde, furfural, p-tolualdehyde, p-anisaldehyde, citral, 2-nitro-
benzaldehyde, and cyclohexanone were obtained from Sigma-Aldrich.
(THF)2MnCl2, 4-fluorobenzaldehyde, and Celite were purchased from
Acros. 2-Naphthaldehyde, 4-bromobenzaldehyde, 4-chlorobenzalde-
hyde, pyridine-3-carboxaldehdye, 3-cyclohexene-1-carboxaldehyde,
acetophenone, and p-toluenesulfonic acid monohydrate were received
from TCI America. Methyl formate, ethyl formate, anisyl formate,
octyl formate, benzyl formate, and isoamyl formate were also
purchased from TCI America. PhSiH3 and 4-cyanobenzaldehyde
were obtained from Oakwood Products. Ethyl acetate was obtained
from VWR while anhydrous Na2SO4 and NaOH were sourced from
Alfa Aesar. All the liquid substrates were scrupulously dried over 4 Å
molecular sieves or distilled if necessary before use. The solid
substrates were recrystallized from diethyl ether or tetrahydrofuran
before use. Ph2PEtPDI was prepared according to literature procedure.21
1
Solution H and 13C nuclear magnetic resonance (NMR) spectra
were recorded in benzene-d61 at room temperature on a Varian 400-
MR NMR spectrometer. All H and 13C NMR chemical shifts (ppm)
are reported relative to Si(CH3)4 using 1H (residual) and 13C chemical
Figure 4. Modified Ojima mechanism proposed for 2-mediated
carbonyl and carboxylate hydrosilylation.
D
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
shifts of the solvent as secondary standards. 31P NMR shifts are
referenced relative to H3PO4 as an external standard. Elemental
analysis was conducted at Robertson Microlit Laboratories Inc.
(Ledgewood, NJ). Although the results lie slightly outside the range
viewed as establishing analytical purity, they are provided to illustrate
the best values obtained to date. Solid-state magnetic susceptibility was
recorded at 25 °C using a Johnson Matthey magnetic susceptibility
balance calibrated with HgCo(SCN)4. Solution state magnetic
susceptibility was determined via Evans method on the Varian 400
MHz NMR spectrometer. Melting point determinations were
performed using a DigiMelt apparatus.
direction and mI1 and mI2 are the magnetic quantum numbers of the
55Mn nuclei.
Fitting of EPR Spectra. To quantitatively compare experimental and
simulated spectra, we divided the spectra into N intervals, i.e., we
treated the spectrum as an N-dimensional vector R. Each component
Rj has the amplitude of the EPR signal at a magnetic field Bj, with j
varying from 1 to N. The amplitudes of the experimental and
simulated spectra were normalized so that the span between the
maximum and minimum values of Rj is 1. We compared the calculated
amplitudes Rcj alc of the signal with the observed values Rj defining a
root-mean-square deviation σ by
X-ray Crystallography. Single crystals suitable for X-ray
diffraction were coated with polyisobutylene oil in the glovebox and
transferred to a glass fiber with Apiezon N grease, which was then
mounted on the goniometer head of a Bruker APEX Diffractometer
equipped with Mo Kα radiation. A hemisphere routine was used for
data collection and determination of the lattice constants. The space
group was identified and the data was processed using the Bruker
SAINT+ program and corrected for absorption using SADABS. The
structure was solved using direct methods (SHELXS) completed by
subsequent Fourier synthesis and refined by full-matrix, least-squares
procedures on [F2]. Crystallographic parameters for complex 2 are
provided in Table S1.
Electron Paramagnetic Resonance Spectroscopy. Instrumen-
tation. Studies were performed at the EPR Facility of Arizona State
University. Continuous wave (CW) EPR spectra were recorded at 107
K using a Bruker ELEXSYS E580 CW X-band spectrometer (Bruker,
Rheinstetten, Germany) equipped with a liquid nitrogen temperature
control system (ER 4131VT). The magnetic field modulation
frequency was 100 kHz with a field modulation of 1 mT peak-to-
peak. The microwave power was 4 mW, the microwave frequency was
9.40 GHz, and the sweep time was 84 s.
σ(p , p , ... , p ) = [ (Rcalc(p , p , ... , p ) − Rjexp)2/N]1/2
∑
j
1
2
n
1
2
n
j
(2)
where the sums are over the N values of j, and p’s are the fitting
parameters that produced the calculated spectrum. For our
simulations, N was set equal to 1024. The EPR spectra were simulated
using EasySpin (v 5.1.10), a computational package developed by Stoll
and Schweiger26 and based on Matlab (The MathWorks, Natick, MA).
EasySpin calculates EPR resonance fields using the energies of the
states of the spin system obtained by direct diagonalization of the spin
Hamiltonian (see eq 1). The EPR fitting procedure used a Monte
Carlo type iteration to minimize the root-mean-square deviation, σ
(see eq 2) between measured and simulated spectra. We searched for
the optimum values of the following parameters: the principal
components of g (i.e., gx, gy, and gz), the ZFS parameters, D and E,
Mni
Mni
the principal components of the hfc tensor AMn (i.e., Ax , Ay , and
i
AzMn) and the peak-to-peak line-widths (ΔBx, ΔBy, and ΔBz).
i
Preparation of (Ph2PEtPDI)MnCl2 (1). A thick-walled glass tube
was charged with 0.269 g of Ph2PEtPDI (0.461 mmol) and 0.124 g of
(THF)2MnCl2 (0.461 mmol) in approximately 20 mL of toluene. The
tube was sealed under nitrogen atmosphere and heated at 125 °C for
120 h. The resulting pale orange slurry was vacuum filtered, and the
solid was washed with toluene (4 × 5 mL) and diethyl ether (3 × 5
mL) to remove any excess ligand. Finally, drying yielded 0.307 g
(94%) of a pale orange solid identified as 1. Elemental analysis for
C37H37N3P2MnCl2: calcd. C, 62.45%; H, 5.24%; N, 5.90%; Found C,
61.81%; H, 5.10%; N, 5.72%. Magnetic moment (Gouy balance, 25
Spin Hamiltonian. The EPR spectrum of [(Ph2PEtPDI)Mn]2 was
analyzed considering that this manganese complex contains two
identical molecules. Each of these molecules was assumed to carry one
net unpaired electron (see Results section), thereby yielding a triplet
state (S = 1) for the manganese complex. Consequently, the EPR data
was fit using a spin Hamiltonian, H, containing the electron Zeeman
interaction with the applied magnetic field Bo, the zero-field
interaction, and the hyperfine coupling (hfc) interactions with two
°C): 6.0 μB. H NMR (CDCl3, 25 °C): No resonances observed. 13C
1
25
5
equivalent 55Mn (I = /2):
NMR (CDCl3, 25 °C): No resonances observed.
Preparation of [(Ph2PEtPDI)Mn]2 (2). A 20 mL scintillation vial
was charged with 4.716 g (23.58 mmol) of Hg0 in 5 mL of THF and
0.027 g (1.18 mmol) of freshly cut Na0 was added. The amalgam was
stirred for 25 min while it became clear. Then, 13.3 μL (0.118 mmol)
of 1,3,5,7-cyclooctatetraene was added and stirred for 5 min while it
turned pale yellow. A 10 mL THF slurry of 1 (0.168 g, 0.236 mmol)
was then added and an instantaneous color change to red was noticed.
The mixture was stirred at room temperature for 6 h. The red solution
was filtered through Celite and THF was removed in vacuo. The
resulting red film was washed with pentane (2 × 5 mL) and dried
again. It was then dissolved in 10 mL of toluene and filtered through a
Celite column. After concentrating the filtrate and layering with diethyl
ether, recrystallization at −35 °C yielded 0.113 g (38%) of red crystals
identified as 2 upon drying. Elemental analysis for C74H74N6P4Mn2
Calcd: C, 69.37%; H, 5.82%; N, 6.56%. Found C, 69.36%; H, 6.35%;
N, 6.21%. Magnetic moment (Evans method, 25 °C): μeff = 3.3 μB. 1H
NMR (benzene-d6, 23 °C): 32.26 (peak width at half height = 514
Hz), 26.23 (153 Hz). 13C NMR (benzene-d6, 25 °C): No resonances
observed. {1H}31P NMR (benzene-d6, 25 °C): −58.49 (487 Hz).
General Method of Aldehyde Hydrosilylation Using 0.05
mol % 2 (0.1 mol % Relative to Mn). In the glovebox, a 20 mL
scintillation vial was charged with 0.0022 g of 2 (0.0017 mmol). To
the catalyst, an equimolar mixture of PhSiH3 (3.40 mmol) and
aldehyde (3.40 mmol) was added. CAUTION: This reaction is
exothermic and vigorous bubbling occurs, which we believe is due to
reactant vaporization. After 2 min, the reaction was exposed to oxygen
to deactivate the catalyst. The resulting colorless solution was filtered
2
Mni
i
H = βS.g.Bo + hS.D.S +
hS.AMn .I
e
∑
(1)
i=1
where S is the electron spin operator, IMn are the nuclear spin
i
operators of the two equivalent 55Mn, D and AMn are the zero-field
i
interaction and hfc tensors, respectively, all in frequency units, g is the
electronic g-tensor, βe is the electron magneton, and h is Planck’s
constant. The so-called zero-field interaction occurs in the absence of
an applied magnetic field because of electron−electron repulsion. For
a triplet state (S = 1) system, the zero-field interaction partially breaks
the degeneracy of the triplet causing the energy of the levels,
2
corresponding to mS = 1, to shift by the term DmS , where D is the
axial zero-field splitting (ZFS) parameter and mS is the magnetic
quantum number of the triplet (0, 1). Additional shifting of the
energy of the mS = 1 doublet is induced by the rhombic zero-field
splitting term, which is characterized by the parameter E. The electron
Zeeman interaction contributes to the Hamiltonian when an external
magnetic field is applied. This interaction is anisotropic and depends
on the relative orientation between the magnetic field and the
molecular axes of the manganese complex. The Zeeman interaction
breaks the remaining degeneracy of the mS = 1 doublet causing an
additional shift given by the term gβeB0mS/h in the energy of these
levels, where g is the g-value. A further energetic consideration is the
contribution of the hfc interactions, which represents the interaction
between the magnetic moment of the unpaired electron system and
the magnetic moments of two equivalent 55Mn nuclei. The hyperfine
interaction is described as first-order by the expression AMn mS(mI1
+
through Celite and analyzed by H NMR spectroscopy to determine
1
i
mI2), where AMn is the hfc interaction along an arbitrary magnetic field
percent conversion. The fractions were recombined and hydrolyzed
i
E
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
with 2 mL of 10% aqueous NaOH solution upon stirring for 2 h at
room temperature. The organic layer was extracted with diethyl ether
(3 × 4 mL) and dried over anhydrous Na2SO4. The solvent was
removed in vacuo at 40 °C to yield the corresponding alcohol.
General Method of Aldehyde Hydrosilylation Using 0.005
mol % 2 (0.01 mol % Relative to Mn). In the glovebox, a 100 mL
round-bottom flask was charged with 0.0014 g of 2 (0.0011 mmol). To
the catalyst, an equimolar mixture of PhSiH3 (22.6 mmol) and
aldehyde (22.6 mmol) was added. CAUTION: This reaction is
exothermic and vigorous bubbling occurs, which we believe is due to
reactant vaporization. After 2 min, the reaction was exposed to oxygen
to deactivate the catalyst. The colorless solution obtained was filtered
REFERENCES
■
(1) Hudlicky, M. Reductions in Organic Chemistry, 2nd ed.; American
́
Chemical Society: Washington, DC, 1996.
(2) (a) Seyden-Penne, J. Reductions by the Alumino- and Borohydrides
in Organic Synthesis, 2nd ed.; Wiley-VCH: New York, 1997. (b) Abdel-
Magid, A. F. Reductions in Organic Synthesis; American Chemical
Society: Washington, DC, 1996.
(3) Rylander, P. N. Catalytic Hydrogenation in Organic Syntheses;
Academic Press: New York, NY, 1979.
(4) Ojima, I. Chem. Org. Silicon Compd. 1989, 2, 1479−1526.
(5) For an influential example, see Ojima, I.; Kogure, T.
Organometallics 1982, 1, 1390−1399.
1
through Celite and analyzed by H NMR spectroscopy to determine
(6) (a) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Sausalito, CA, 1987. (b) Hartwig, J. F.
Organotransition Metal Chemistry − From Bonding to Catalysis;
University Science Books: Sausalito, CA, 2010.
the percent conversion. The fractions were recombined and
hydrolyzed with 2 mL of 10% aqueous NaOH solution upon stirring
for 2 h at room temperature. The organic layer was extracted with
diethyl ether (3 × 4 mL) and dried over anhydrous Na2SO4. The
solvent was removed in vacuo at 40 °C to yield the corresponding
alcohol.
(7) Leading Fe carbonyl hydrosilylation catalysts according to TOF:
(a) Yang, J.; Tilley, T. D. Angew. Chem., Int. Ed. 2010, 49, 10186−
10188. (b) 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.
General Method of Formate Dihydrosilylation Using 0.01
mol % 2 (0.02 mol % Relative to Mn). In the glovebox, a 100 mL
round-bottom flask was charged with 0.002 g of 2 (0.002 mmol). To
the catalyst, an equimolar mixture of PhSiH3 (20.0 mmol) and formate
(20.0 mmol) was added. CAUTION: This reaction is exothermic and
vigorous bubbling occurs, which we believe is due to reactant
vaporization. After 30 min, the reaction was exposed to oxygen to
deactivate the catalyst. The colorless solution obtained was filtered
(8) Leading Co carbonyl hydrosilylation catalysts according to TOF:
(a) Niu, Q.; Sun, H.; Li, X.; Klein, H.-F.; Florke, U. Organometallics
̈
2013, 32, 5235−5238. (b) Nesbit, M. A.; Suess, D. L. M.; Peters, J. C.
Organometallics 2015, 34, 4741−4752.
1
through Celite and analyzed by H NMR spectroscopy to determine
(9) Leading Ni carbonyl hydrosilylation catalysts according to TOF:
(a) Postigo, L.; Royo, B. Adv. Synth. Catal. 2012, 354, 2613−2618.
(b) Bheeter, L. P.; Henrion, M.; Brelot, L.; Darcel, C.; Chetcuti, M. J.;
Sortais, J.-B.; Ritleng, V. Adv. Synth. Catal. 2012, 354, 2619−2624.
(10) Leading Cu carbonyl hydrosilylation catalysts according to
TOF: (a) Lipshutz, B. H.; Noson, K.; Chrisman, W.; Lower, A. J. Am.
Chem. Soc. 2003, 125, 8779−8789. (b) Wu, J.; Ji, J.-X.; Chan, A. S. C.
Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 3570−3575.
the percent conversion. The fractions were recombined and
hydrolyzed with 2 mL of 10% aqueous NaOH solution upon stirring
for 2 h at room temperature. The organic layer was extracted with
diethyl ether (3 × 4 mL) and dried over anhydrous Na2SO4. The
solvent was removed in vacuo at 40 °C to yield the corresponding
alcohol.
(11) Leading Zn carbonyl hydrosilylation catalysts according to
TOF: (a) Rit, A.; Zanardi, A.; Spaniol, T. P.; Maron, L.; Okuda, J.
Angew. Chem., Int. Ed. 2014, 53, 13273−13277. (b) Sattler, W.;
Ruccolo, S.; Chaijan, M. R.; Allah, T. N.; Parkin, G. Organometallics
2015, 34, 4717−4731.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00423.
(12) (a) Trovitch, R. J. Synlett 2014, 25, 1638−1642. (b) Valyaev, D.
A.; Lavigne, G.; Lugan, N. Coord. Chem. Rev. 2016, 308, 191−235.
(13) Manganese carbonyl reduction catalysts that utilize H2 or
boranes have recently been described. See (a) Zhang, G.; Zeng, H.;
Wu, J.; Yin, Z.; Zheng, S.; Fettinger, J. C. Angew. Chem., Int. Ed. 2016,
55, 14369−14372. (b) Elangovan, S.; Topf, C.; Fischer, S.; Jiao, H.;
Spannenberg, A.; Baumann, W.; Ludwig, R.; Junge, K.; Beller, M. J.
Am. Chem. Soc. 2016, 138, 8809−8814. (c) Kallmeier, F.; Irrgang, T.;
Dietel, T.; Kempe, R. Angew. Chem., Int. Ed. 2016, 55, 11806−11809.
Compound characterization, metrical parameters for 2
and detailed hydrosilylation procedures (PDF)
Accession Codes
CCDC 1552985 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_
request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: +44 1223 336033.
(d) Zirakzadeh, A.; de Aguiar, S. R. M. M.; Stoger, B.; Widhalm, M.;
̈
Kirchner, K. ChemCatChem 2017, 9, 1744−1748. (e) Widegren, M. B.;
Harkness, G. J.; Slawin, A. M. Z.; Cordes, D. B.; Clarke, M. L. Angew.
Chem., Int. Ed. 2017, 56, 5825−5828. (f) Vasilenko, V.; Blasius, C. K.;
Wadepohl, H.; Gade, L. H. Angew. Chem., Int. Ed. 2017, 56, 8393−
8397.
AUTHOR INFORMATION
Corresponding Author
*E-mail: ryan.trovitch@asu.edu.
■
(14) Cavanaugh, M. B.; Gregg, B. T.; Cutler, A. R. Organometallics
1996, 15, 2764−2769.
(15) This paper describes the conversion of esters into ethers via
deoxygenation as opposed to ester dihydrosilylation to yield a mixture
of silyl ethers. Mao, Z.; Gregg, B. T.; Cutler, A. R. J. Am. Chem. Soc.
1995, 117, 10139−10140.
ORCID
Ryan J. Trovitch: 0000-0003-4935-6780
Notes
The authors declare no competing financial interest.
(16) (a) 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.
(b) Son, S. U.; Paik, S.-J.; Chung, Y. K. J. Mol. Catal. A: Chem. 2000,
151, 87−90. (c) Zheng, J.; Chevance, S.; Darcel, C.; Sortais, J.-B.
Chem. Commun. 2013, 49, 10010−10012. (d) Zheng, J.; Elangovan, S.;
ACKNOWLEDGMENTS
■
This material is based upon work supported by the National
Science Foundation under Grant No. 1651686. Acknowledge-
ment is also made to the Donors of the American Chemical
Society Petroleum Research Fund for support of this research.
We acknowledge Dr. Brian R. Cherry for NMR assistance.
́
Valyaev, D. A.; Brousses, R.; Cesar, V.; Sortais, J.-B.; Darcel, C.; Lugan,
N.; Lavigne, G. Adv. Synth. Catal. 2014, 356, 1093−1097. (e) Valyaev,
D. A.; Wei, D.; Elangovan, S.; Cavailles, M.; Dorcet, V.; Sortais, J.-B.;
Darcel, C.; Lugan, N. Organometallics 2016, 35, 4090−4098.
F
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(f) Vijjamarri, S.; Chidara, V. K.; Du, G. ACS Omega 2017, 2, 582−
591.
(17) Chidara, V. D.; Du, G. Organometallics 2013, 32, 5034−5037.
(18) Mukhopadhyay, T. K.; Flores, M.; Groy, T. L.; Trovitch, R. J. J.
Am. Chem. Soc. 2014, 136, 882−885.
(19) Mukhopadhyay, T. K.; Rock, C. L.; Ashley, D. C.; Hong, M.;
Groy, T. L.; Baik, M.-H.; Trovitch, R. J. J. Am. Chem. Soc. 2017, 139,
4901−4915.
(20) Ghosh, C.; Mukhopadhyay, T. K.; Flores, M.; Groy, T. L.;
Trovitch, R. J. Inorg. Chem. 2015, 54, 10398−10406.
(21) Ben-Daat, H.; Hall, G. B.; Groy, T. L.; Trovitch, R. J. Eur. J.
Inorg. Chem. 2013, 2013, 4430−4442.
(22) Bart, S. C.; Chlopek, K.; Bill, E.; Bouwkamp, M. W.; Lobkovsky,
E.; Neese, F.; Wieghardt, K.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128,
13901−13912.
(23) (a) Khusniyarov, M. M.; Harms, K.; Burghaus, O.;
Sundermeyer, J. Eur. J. Inorg. Chem. 2006, 2006, 2985−2996.
(b) Muresan, N.; Chlopek, K.; Weyhermuller, T.; Neese, F.;
̈
Wieghardt, K. Inorg. Chem. 2007, 46, 5327−5337.
(24) The leading TOF for each metal previously reported to mediate
carboxylate dihydrosilylation via selective acyl C−O cleavage is
provided: (a) For Ti (TOF = 33 h−1): Barr, K. J.; Berk, S. C.;
Buchwald, S. L. J. Org. Chem. 1994, 59, 4323−4326. (b) For V (TOF
= 0.8 h−1): Pehlivan, L.; Met
́
ay, E.; Laval, S.; Dayoub, W.; Delbrayelle,
D.; Mignani, G.; Lemaire, M. Eur. J. Org. Chem. 2011, 2011, 7400−
7406. (c) For Mn, see ref 19. (d) For Fe (TOF as high as 100 h−1),
see ref 7b. (e) For Zn (TOF = 12.5 h−1): Mimoun, H. J. Org. Chem.
1999, 64, 2582−2589. (f) For Mo (TOF = 1.0 h−1) Fernandes, A. C.;
Romao, C. C. J. Mol. Catal. A: Chem. 2006, 253, 96−98. (g) For Ru
̃
(TOF = 33 h−1, based on Ru) Matsubara, K.; Iura, T.; Maki, T.;
Nagashima, H. J. Org. Chem. 2002, 67, 4985−4988. (h) For Rh (TOF
= 16.7 h−1) Ohta, T.; Kamiya, M.; Kusui, K.; Michibata, T.;
Nobutomo, M.; Furukawa, I. Tetrahedron Lett. 1999, 40, 6963−
6966. (i) For Pd (TOF = 2.8 h−1) Nakanishi, J.; Tatamidani, H.;
Fukumoto, Y.; Chatani, N. Synlett 2006, 2006, 869−872.
(25) Weil, J. A.; Bolton, J. R. Electron paramagnetic resonance:
Elementary theory and practical applications; Wiley: Hoboken, NJ, 2007.
(26) Stoll, S.; Schweiger, A. J. Magn. Reson. 2006, 178, 42−55.
G
DOI: 10.1021/acs.organomet.7b00423
Organometallics XXXX, XXX, XXX−XXX