Efficient Hydrosilylation of Acetophenone with a New Anthraquinonic Amide-Based Iron Precatalyst

Article abstract of DOI:10.1021/acs.organomet.6b00765

A new iron complex based on a noninnocent anthraquinonic ligand has been synthesized and fully characterized through multiple techniques, including NMR, X-ray crystallography, mass spectrometry, and cyclic voltammetry. Exposure of ketone to that complex i

Full text of DOI:10.1021/acs.organomet.6b00765

Article
pubs.acs.org/Organometallics
Efficient Hydrosilylation of Acetophenone with a New
Anthraquinonic Amide-Based Iron Precatalyst
†
‡
†
§
́
Alvaro Raya-Baron, Manuel A. Ortuno, Pascual Ona-Burgos, Antonio Rodríguez-Dieguez,
́
́
̃
̃
Robert Langer,∥ Christopher J. Cramer,*,‡ Istemi Kuzu,*,∥ and Ignacio Fernandez*,†,⊥
́
†Department of Chemistry and Physics, ceiA3, Universidad de Almería, Ctra. Sacramento, s/n, E-04120 Almería, Spain
‡Department of Chemistry, Chemical Theory Center, and Supercomputing Institute, University of Minnesota, Minneapolis,
Minnesota 55455-0431, United States
§Department of Inorganic Chemistry, Faculty of Science, University of Granada, 18070 Granada, Spain
∥Fachbereich Chemie, Philipps-Universitat Marburg, Hans-Meerwein-Straße 4, 35032 Marburg, Germany
̈
⊥BITAL, Research Centre for Agricultural and Food Biotechnology, Ctra. Sacramento, s/n, E-04120 Almería, Spain
S
* Supporting Information
ABSTRACT: A new iron complex based on a noninnocent anthraquinonic ligand
has been synthesized and fully characterized through multiple techniques,
including NMR, X-ray crystallography, mass spectrometry, and cyclic voltammetry.
Exposure of ketone to that complex in the presence of (EtO)2MeSiH affords the
corresponding silylated alcohol. Loadings as low as 0.25 mol % afford excellent
yields at room temperature. Quantum−chemical analysis of the catalytic
mechanism supports activation of the precatalyst to form a Fe−hydride
intermediate, followed by ketone reduction and σ-bond metathesis.
computational mechanistic studies on the hydrosilylation of
acetophenone catalyzed by the new complex are also presented.
The ligand containing an N-heterocyclic chelating element
has been synthesized from common, inexpensive starting
materials in high yields, from the reaction of chloroanthraqui-
none with 2-picolylamine at 150 °C over short reaction times.
The isolated yield is 40% (see Supporting Information for
NMR, mass spectrometry, IR, and X-ray crystallographic data).
A key identifying feature in the 1H NMR spectra of the ligand is
the presence of a broad triplet NH resonance at δH 10.50 ppm,
due to hydrogen-bonding with the CO moiety, together with
the doublet associated with the methylene bridge (δH 4.84
he hydrosilylation of carbonyl groups is an important
Tsynthetic transformation and one that is often used to
evaluate specific catalysts for potential further utility if
engineered to accomplish CO2 reduction.1 Hydrosilylation
combines reduction of the carbonyl functionality with alcohol
protection in a single step, where the most common and active
catalysts are based on precious metals.2 In the past decade,
much effort has been dedicated to the development of catalysts
based on earth-abundant metals, which has opened unprece-
dented routes for molecular functionalization.3 Iron, in
particular, has been exploited because of its benign environ-
mental impact, high abundancy in nature, and low price, and
has emerged as one of the most noteworthy substitutes in the
catalyzed hydrosilylation.4−6 From a mechanistic perspective,
there are few contributions dealing with this transformation,
mostly on low-spin iron(II) hydride catalysts.3d,4a,e,f,i,5−8 Very
recently, Gade and co-workers have given insights in the
mechanism of chiral iron carboxylate precatalysts,5,9 identifying
a rate-determining σ-bond metathesis step of the complex with
the silane, subsequent coordination of a ketone to the iron
hydride complex, and finally an insertion of the ketone into the
Fe−hydride bond to regenerate the active species.
1
ppm). H,15N HMQC experiments optimized for 1/2JNH of
83.3 ms (Figure S15) permitted the assignment of two
nitrogens at δN 77.9 (NH) and 307.0 ppm for the picolyl
and pyridine units, respectively. The former interacts
exclusively with the NH proton, while the latter shows cross
peaks with its vicinal proton and the methylenic moiety. The X-
ray crystal structure of 1 (Figure 1), grown by hexane diffusion
in a dicloromethane solution, confirms the chelating potential
of the ligand given the intramolecular hydrogen bond between
NH and CO (quinone) of 2.629 Å (Figures 1 and S26). This
Herein, we report the synthesis and characterization of a new
anthraquinonic amide-base complex that includes two nitrogen
atoms in the tridentate ligand. Preliminary results and
Received: October 1, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a stable complex not prone to dissociation and/or aggregation
1
on the NMR time scale. Despite their paramagnetism, the H
NMR spectra of complexes 2 and 3 reported in this work are
informative and have been partially assigned.
Complex 2 proved to be stable in THF solution over several
weeks without any sign of decomposition. The THF-d8 1H
NMR spectrum of 2 at 23 °C exhibits a number of resonances
consistent with a paramagnetic C2v-symmetric iron complex
having resonances shifted over a 200 ppm range.11 Variable
temperature 1H NMR spectra and a list of diagnostic signals are
provided in the Supporting Information. In general, the in-
plane hydrogens on the chelate exhibit relatively large
deviations from their diamagnetic reference values. In THF-
d8, the pyridine protons are observed as rather broad downfield
Figure 1. Molecular structure of ligand 1 determined from single-
crystal X-ray diffraction. Thermal ellipsoids are shown at the 30%
probability level.
peaks at δH 154.4 (W1/2 = 3019 Hz) and 136.8 ppm (W1/2
=
2375 Hz). The three closer to the iron anthraquinone protons
resonate at δH 85.4, 57.4, and 35.9 ppm, while the most remote
appeared as sharper singlets located at δH 37.7 (W1/2 = 43 Hz),
27.1 (W1/2 = 84 Hz), and −28.2 ppm (W1/2 = 98 Hz).
Transient complex 3 shows a number of resonances consistent
with a Cs paramagnetic complex with higher and lower
frequency signals located at δH 189.2 and −34.1 ppm,
interaction produces a slightly longer carbonyl bond (C9−O1,
1.238(2) Å) than that found for the non-hydrogen-bonded case
(C10−O2 1.227(2) Å). In the crystal network, anthraquinone
moieties of 1 are oriented in a face-to-face manner with an
interplanar distance of ca. 3.4 Å, establishing π-stacking
arrangements (Figures S27−28). This interaction induces
architecturally controlled self-assembly, leading to molecular
aggregates in the solid state, as has been described previously.10
The reaction of 1 with 0.5 equiv of Fe(HMDS)2 in THF led
to quantitative formation of complex 2 (Scheme 1).
1
respectively. Accordingly, the H NMR spectrum in THF-d8
at 23 °C exhibits diagnostic resonances with large isotropic
shifts for the in-plane hydrogens (Figure S17). The Evans
method was used to calculate the solution magnetic
susceptibility values, χM, for complex 2, giving rise to a number
of unpaired electrons (n = 4) consistent with a high-spin Fe(II)
electronic configuration (see Supporting Information and
Figure S25). The FT-IR spectra (Figure S21) showed two
CO stretching bands at 1641 and 1580 cm−1, indicating the
unsymmetrical coordination of the two quinone oxygens. As far
as we are aware, this is the first example of an anthraquinonic
amide-based iron complex described to date.
Scheme 1. Synthesis of Anthraquinonic Iron Complexes 2
a
and 3
Purple single crystals were obtained from layering with n-
pentane a THF solution of complex 2 at room temperature.
The same compound was obtained when using dichloro-
methane instead of THF. The two crystallized compounds (2
in THF, and 2b in dicloromethane) are essentially the same but
with different crystallization solvents included in the lattice (see
Supporting Information). The geometry about Fe can best be
described as a distorted octahedron with two tridentate ligands
bound to the metal with four nitrogens (two pyridinic and two
amidic) and two weakly interacting carbonyl oxygens (Figure
2). Bond angles of 170.50(6), 78.70(5), and 94.58(5)°, are
observed for N(1)−Fe(1)−N(3), N(1)−Fe(1)−N(2), and
O(1)−Fe(1)−O(3) respectively (Figure 2).
Inspection of the anthraquinone chelate carbonyl CO
distances reveals considerable elongation [1.268(2) and
1.2646(19) Å] relative to that of the uncomplexed carbonyls
[1.232(2) and 1.228(2) Å], consistent with NMR and IR data.
Corresponding bond distances and angles are given in Tables
S2 and S3. The positive ion mode ESI mass spectrum reveals a
parent ion corresponding to [2]+ m/z 682.1296 (Figure S23),
confirming its full integrity in the gas phase. The observed
isotope pattern matches well with that calculated for the exact
mass of [C40H26FeN4O4] (Figures S23 and 24).
a
See the Supporting Information for more details. We tentatively
assume that two THF molecules complete the coordination sphere of
iron in complex 3.
Interestingly, when the reaction between ligand 1 and
Fe(HMDS)2 is performed using a 1:1 molar ratio (ligand:Fe
precursor), complex 3 is initially obtained (Scheme 1), which
progressively disappears forming again complex 2. NMR
monitoring of the 1:1 stoichiometry reaction (Figure S18)
indicated that after 12 h under these conditions quantitative
conversion to complex 2 is achieved.
Thus, there is a thermodynamic equilibrium between
complexes 2 and 3, with the former being the thermodynami-
cally more stable species. For ligand/iron stoichiometries
The redox properties of complex 2 were investigated by
cyclic voltammetry in dichloromethane. The redox potentials
are given versus the standard Fc/Fc+ couple. The first reversible
one-electron reduction of the anthraquinone moiety in complex
2 to the corresponding radical anion was observed at ca. −0.85
V vs Fc/Fc+ couple. The irreversible reduction wave at −1.37 V
1
greater than 2:1, e.g., 3:1 or 4:1, the H NMR spectra showed
the coexistence of complex 2 and free ligand in agreement with
B
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
Table 1. Hydrosilylation of Acetophenone Catalyzed by 2
entry
Fe prec.
1 (equiv)
silane
t (h)
conv. (%)
52
b
1
2
(EtO)2MeSiH
(EtO)2MeSiH
Ph2SiH2
1
24
1
e
2
3
4
5
2
80 (90)
2
3
2
Ph2SiH2
24
62
1
45
96
77
50
0
2
Ph2SiH2
c
6
c
Fe(OAc)2
2.2
2.2
(EtO)2MeSiH
(EtO)2MeSiH
(EtO)2MeSiH
(EtO)2MeSiH
(EtO)2MeSiH
(EtO)2MeSiH
Ph2SiH2
7
FeCl2
1
8
9
Fe(OAc)2
24
24
1
FeCl2
0
d
10
2
50
85
72
96
0
d
11
2
24
24
62
24
24
24
24
d
12
2
d
13
2
Ph2SiH2
14
2
Et3SiH
c
Figure 2. Molecular structure of complex 2 determined from single-
crystal X-ray diffraction. Thermal ellipsoids are shown at the 30%
probability level. Hydrogen atoms are omitted for clarity.
15
Fe(OAc)2
2
2.2
2.2
Et3SiH
0
16
Ph3SiH
0
c
17
Fe(OAc)2
Ph3SiH
0
a
Reaction conditions unless stated otherwise: acetophenone (2
in the cyclic voltammogram of 2 is assignable to a second
reduction of the anthraquinone-based ligand in 2. In
comparison to those of the uncoordinated ligand (1) the
observed redox couples of the anthraquinone moiety in 2 are
shifted by +0.58 and +0.65 V upon coordination to iron(II)
(Figure 3). While the voltammogram of 1 exhibits distinct
mmol), hydrosilane (2.5 mmol), Fe precursor (0.25 mol %), LiHMDS
(base), THF (6 mL), room temperature. Quenched with 1 M HCl
(aqueous). Conversions determined by crude 1H NMR. See
b
Supporting Information. When the reaction is performed in absence
c
of THF, the same conversion is obtained. Two equivalents of
d
e
LiHMDS were employed. Addition of LiOAc (2 equiv). In
parentheses are shown the conversions when using 0.5 mol % of
catalyst.
the catalytic reactions proceed quantitatively, although the use
of Ph2SiH2 requires about three times longer to proceed with
excellent yields. When complex 2 is prepared in situ from
Fe(OAc)2 in the reaction vessel (entry 6), the conversion is
notably higher than when using 2 in isolated form. Under these
conditions, the presence of the amide ligand is required in
order to observe catalysis (entries 8 and 9). The hydrosilylation
of ketones catalyzed by a combination of Fe(OAc)2 with
various N-coordinating ligands has been reported previously by
Nishiyama and Furuta6 but with reaction temperatures of 65 °C
and loadings of catalysts above 5%. The substitution of the
acetate precursor by FeCl2 (entry 7) produced the same
catalytic outcome as when using complex 2 in isolated form.
The higher yields from in situ generated catalyst (entry 1 vs
entry 6) encouraged us to employ complex 2 in the presence of
LiOAc as an initiator. The yields are shown in entries 10−13
and showed no significant improvement for either EtO2MeSiH
or Ph2SiH2. Neither triethylsilane nor triphenylsilane displayed
any reactivity under isolated or in situ-generated catalytic
conditions (entries 14−17). We attribute this to the lesser
hydridic characters for these less Si−H polarized reagents.12
The estimated TOF values for complex 2 were 3.5 min−1
(entry 1, Table 1) and 5.1 min−1 (entry 6, Table 1). To place
these activities in context, it is fair to evoke the calculated TOF
values for some of the most active iron complexes in the same
transformation. In this sense, special mention is given to that
reported by Ruddy et al. (393 min−1)4g or that described by
Yang and Tilley (∼10 min−1).4d A comparison of prominent
first-row metal catalysts has been provided by Trovitch and co-
workers.13 Importantly, in our case no special neutral donor
ligand or activator is required. It is noteworthy that all reactions
were conducted at room temperature, illustrating the useful
activity of the catalysts described herein.
Figure 3. Cyclic voltammogram of complex 2 in CH2Cl2 with different
scan directions (1 mM, scan rate 100 mV/s, Pt//0.1 M nBu4NPF6//
Pt).
redox couples at 0.80 and 1.10 V, which are in line with amine-
and pyridine-oxidation, respectively, complex 2 shows an
irreversible oxidation process at 1.38 V that we assign to
FeII/FeIII-couple.
We were glad to find that complex 2 could be prepared in
THF from the reaction of Fe(OAc)2 (or alternatively FeCl2),
ligand 1, and LiHMDS in a ratio 1:2.2:2, respectively (see
Supporting Information). These routes were important in order
to use a much more convenient iron source in catalysis. With
the new iron complex fully characterized in solution and the
solid state, we tested its efficiency for the hydrosilylation of
acetophenone (Table 1). Stirring the reaction mixture at room
temperature for 1 h followed by workup afforded the
corresponding alcohol in 52 and 3% yields, when using silanes
EtO2MeSiH or Ph2SiH2, respectively (entries 1 and 3).
Interestingly, at longer reaction times (entries 2, 4, and 5)
C
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
We now turn to electronic structure calculations to
investigate the reaction mechanism. The hydrosilylation of
acetophenone described in entry 6 of Table 1 has been
modeled at the density functional theory (DFT) level (see
Computational Details). Starting from the precatalyst 2, we
have removed one anthraquinonic ligand to create available
coordination sites to catalyze reactivity (Figure 4a). We have
activation of the catalyst (Scheme 2a). We start from a van der
Waals complex of [Fe(κ2-AcO)] and silane 4 (−7.9 kcal
mol−1), which undergoes σ-bond metathesis via TS4−5 (18.4
kcal mol−1) to generate the hydride 5.
In the subsequent hydrosilylation catalytic cycle (Scheme
2b), ketone coordination takes place via TS5−6 (10.5 kcal
mol−1), followed by insertion via TS6−7 (12.1 kcal mol−1) (see
isomer in Figure S41a), which reduces the carbonyl group
producing alkoxide 7 (−15.8 kcal mol−1). Next, silane
coordinates to iron via TS8−9 (−9.3 kcal mol−1) to form a
5-coordinate silicon atom in 9 (−8.1 kcal mol−1). Further
conformational rearrangements exchanging the methoxy and
alkoxide positions generates complex 10 that is in equilibrium
with a higher energy agostic-like complex 11 via TS10−11 (0.6
kcal mol−1) (see isomer in Figure S41b). The latter can
undergo σ-bond metathesis via TS11−5 (−4.0 kcal mol−1) to
regenerate the hydride 5 and release the protected alcohol.
In the above mechanistic scenario, the anthraquinonic ligand
is only a spectator, but we did consider its possible participation
in the proposed reaction mechanism. As shown in Scheme 2b,
the transition state TS6−7 features a hydride insertion into the
carbonyl of the substrate at 12.1 kcal mol−1. In this line, the
anthraquinonic ligand also contains a carbonyl group that might
be reduced. Computed transition state TS6−L that describes
the hydride insertion into the anthraquinonic carbonyl is found
at 27.7 kcal mol−1 above 5, i.e., this step requires 15.6 kcal
mol−1 more than the hydride insertion into the substrate via
TS6−7. Although the process cannot be completely ruled out,
it is unfavorable and nonproductive toward catalysis.
Figure 4. (a) Active species and (b) general reaction mechanism for
Fe(II)-catalyzed hydrosilylation of ketones.
also considered one additional explicit THF molecule solely to
complete the coordination sphere, since no major role is
expected (Table 1, entry 1, footnote b). For the silane, ethoxy is
replaced by methoxy for computational efficiency. All Fe(II)
complexes are high-spin (quintet).14 All energies correspond to
298 K Gibbs free energies in THF solution in kcal mol−1.
In line with previous studies,4e,i,7,9,12,15 we propose the
participation of an Fe(II)−hydride as an active species in the
catalytic cycle (Figure 4b), which proceeds as (i) coordination
and reduction of the ketone, (ii) silane activation, and (iii) Fe−
hydride regeneration and product release (see Lewis acid
mechanism in Figure S40).
The overall Gibbs free energy profile is shown in Figure 5.
Silane activation and Fe−hydride regeneration taken in
The computed reaction mechanism is shown in Scheme 2,
where separated Fe−hydride 5 and relevant reagents are taken
as the zero of free energy. The first step of the reaction entails
a
Scheme 2. Computed Reaction Mechanisms
Figure 5. Gibbs energies along reaction coordinate for Fe-catalyzed
hydrosilylation of acetophenone.
sequence comprise σ-bond metathesis. The associated free
energy of activation for this process (ΔG⧧ = 16.4 kcal mol−1
relative to 7) is higher than that of ketone reduction (ΔG⧧ =
12.1 kcal mol−1 relative to 5). In the analogous process
reported by Bleith and Gade9 for a different iron-based catalyst
reacting with acetophenone and (MeO)2MeSiH, the computed
free energy of activation for σ-bond metathesis is similar (ΔG⧧
= 12.0 kcal mol−1 compared to ΔG⧧ = 11.8 kcal mol−1 (TS11−
5 relative to 7)), but the free energy of activation computed for
ketone reduction was considerably lower (ΔG⧧ = 6.9 kcal
mol−1 compared to ΔG⧧ = 12.1 kcal mol−1 (TS6−7 relative to
5)), presumably owing to the greater Lewis acidity of their iron
complex facilitating the first step. This suggests that next-
generation design within the context of our own ligand system
might be to substitute the pyridyl ring with electron-
a
(a) Catalyst activation and (b) hydrosilylation of acetophenone.
ΔGTHF in kcal mol−1.
D
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
gel column using ethyl acetate/hexanes (1:1) as eluent. Yield: 1.23 g
(40%). Suitable crystals for X-ray diffraction were grown by layering a
solution of 1 in dichloromethane with hexane at 5 °C. Elemental
analysis calcd (%) for C20H114N2O2: C, 76.42; H, 4.49; N, 8.91; found:
C, 76.44; H, 4.60; N, 8.73. H NMR (600.13 MHz, CDCl3, δ, ppm)
10.49 (1H, bs, NH), 8.76 (1H, ddd, J = 1.2 Hz, 1.8 Hz, 4.8 Hz, H20),
8.38 (1H, dd, J = 0.6 Hz, 7.8 Hz, H8), 8.34 (1H, dd, J = 0.6 Hz, 7.8
Hz, H5), 7.86 (1H, ddd, J = 1.2 Hz, 7.2 Hz, 7.2 Hz, H7), 7.81 (1H,
ddd, J = 1.2 Hz, 7.2 Hz, 7.2 Hz, H6), 7.78 (1H, ddd, J = 1.8 Hz, 7.8
Hz, 7.8 Hz, H18), 7.72 (1H, dd, J = 1.2 Hz, 7.2 Hz, H4), 7.59 (1H, dd,
J = 8.4 Hz, 8.4 Hz, H3), 7.46 (1H, d, J = 7.8 Hz, H17), 7.33 (1H, dd, J
= 5.4 Hz, 7.2 Hz, H19), 7.11 (1H, dd, J = 0.6 Hz, 8.4 Hz, H2), 4.84
(2H, d, J = 5.4 Hz, H15). 13C NMR (150.92 MHz, CDCl3, δ, ppm)
185.4 (CO, C9), 183.7 (CO, C10), 157.6 (C16), 151.3 (C1), 149.4
(C20), 137.2 (C18), 135.4 (C3), 134.9 (C12), 134.7 (C14), 134.0
(C7), 133.1 (C6), 133.1 (C11), 126.9 (C8), 126.8 (C5), 122.5 (C19),
121.2 (C17), 118.3 (C2), 116.3 (C4), 113.8 (C13), 48.6 (CH2, C15).
withdrawing functionality to similarly accelerate ketone
reduction.
The relatively small computed free energy of activation for
the catalytic cycle involving 5 is consistent with the observed
efficiency and mild conditions for the experimental process. We
note that the activation free energy for initial catalyst activation
(ΔG⧧ = 26.3 kcal mol−1 from 4 through TS4−5)16 is actually
larger than the turnover-limiting step for catalysis; thus, an
induction period prior to catalysis is expected, which has been
reported by Bleith and Gade as well.9 Transition-state theory
based on the computed results predicts a TOF of 0.42 min−1,17
which is within an order of magnitude of the experimental
estimate of 5.1 min−1 (entry 6 Table 1).
We may also use computation to predict the kinetic isotope
effect (KIE)18 for catalyst activation (from 4 through TS4−5),
ketone reduction (from 5 through TS6−7), and σ-bond
metathesis (from 7 through TS11−5). The computed H/D
KIE values are 1.88, 1.04, and 1.51, respectively. Whereas the
H/D KIE for the catalyst activation step is similar to previous
experiments using carboxylate precatalysts,9 the values for
ketone reduction and σ-bond metathesis are somewhat
smaller.19
In summary, readily accessed, well-defined anthraquinone
amide-based iron complexes can be isolated (or prepared in
situ) as efficient catalysts for the hydrosilylation of carbonyls.
The method is usable under very mild conditions and using
loadings of catalyst as low as 0.25 mol % with no need for
special neutral donors or activators. Calculations suggest a large
activation free energy for the catalyst preactivation step to form
the initial Fe−hydride, after which ketone reduction and σ-
bond metathesis processes occur readily. Further derivatization
of the ligand and their use in more challenging catalysis are in
progress.
+
ESI-MS calcd (m/z) for C20H15N2O2 : 315.1128; found: 315.1122 [M
+ H]+. IR (KBr) v (cm−1) 3269m, 2923m, 2853m, 1668s, 1635s,
1591s, 1505s, 1481s, 1431s, 1409s, 1298s, 1270s, 1230s, 1001m, 707s.
Synthesis and Analytical Data for Complex 2. A solution of
Fe(HMDS)2 (56.4 mg, 0.15 mmol) in THF (10 mL) was added to a
suspension of 1 (94.2 mg, 0.30 mmol) in THF (10 mL). The resulting
solution was stirred at room temperature for 20 min; then, all the
volatiles were removed under reduced pressure. The pure product was
obtained as dark purple crystals by layering a THF solution with n-
pentane. Yield: 79 mg (77%). Crystals can be also grown from layering
with n-pentane a CH2Cl2 solution of complex 2 at room temperature.
Complex 2 can be also synthesized using FeCl2 as iron(II) precursor.
Thus, a solution of LiHMDS (56.4 mg, 0.32 mmol) in THF (5 mL)
was added to a solution of 1 (100 mg, 0.32 mmol) in THF (15 mL).
The resulting blue solution was poured upon a suspension of FeCl2
(40.5 mg, 0.16 mmol) in THF (5 mL). Then, triethylamine (177 μL,
0.64 mmol) was added and the mixture stirred during 30 min. After
removing all the volatiles under vacuum, dichloromethane was added
in order to dissolve 2 and remove LiCl by filtration. Pure crystalline
product was obtained after layering n-pentane upon a dichloromethane
solution. Elemental analysis calcd (%) for C40H26FeN4O4·(THF): C,
1
EXPERIMENTAL SECTION
70.03; H, 4.54; N, 7.42; found: C, 69.06; H, 4.52 N, 7.69. H NMR
■
(300.13 MHz, THF-d8, δ, ppm) 154.4 (1H, bs), 136.8 (1H, bs), 85.4
(1H, bs), 57.4 (1H, bs), 37.7 (1H, s), 35.9 (1H, bs), 30.1 (1H, bs),
27.1 (1H, s), 7.4 (1H, s), 7.1 (1H, s), −1.6 (1H, s), −28.2 (1H, s),
General Information. All experiments were performed under an
inert atmosphere of N2 or argon using standard Schlenk techniques or
a glovebox. Deuterated solvents were degassed and dried over
activated molecular sieves prior to use. DMSO was dried by distillation
from anhydrous CaH2 and then stored over activated 3 Å molecular
sieves. 2-Picolylamine was distilled from KOH under reduced pressure
before use. 1-Chloroanthraquinone was purchased from Aldrich and
used as received. THF and hexane were dried and degassed in a
Solvent Purification System. Fe(HMDS)2 was synthesized by a
method previously described in literature.20 Pentane used for
crystallization was dried by distillation from Na/K alloy and stored
over 4 Å molecular sieves. LiHMDS, FeCl2, and Fe(OAc)2 were
purchased from ABCR and used as received. All hydrosilanes
employed in the catalytic runs were purchased from Acros and used
without further purification. Acetophenone was distilled from
anhydrous MgSO4 under reduced pressure and collected over
activated molecular sieves, then degassed. Triethylamine was distilled
from CaH2, degassed, and stored over activated molecular sieves.
NMR spectra were measured via Bruker Avance III 300, Bruker
Avance III 500, or Bruker Avance III 600 spectrometers. IR spectra
were recorded in a FT-IR Bruker Alpha spectrometer. Mass spectra
were acquired with a LTQ-FT Ultra mass spectrometer (Thermo
Fischer Scientific). Elemental analyses (EA) were performed on a
Elementar vario EL cube in the CHN mode.
+
−31.9 (1H, bs). ESI-MS calcd (m/z) for C40H26FeN4O4 : 682.1303;
found: 682.1296 [M]+. IR (ATR) v (cm−1) 3066w, 2954w, 2925w,
2852w, 1641s, 1580s, 1522m, 1474m, 1410s, 1347s, 1304s, 1246s,
1227s, 994s, 704s, 487s.
Observation of Complex 3. A solution of 1 (43.2 mg, 0.137
mmol) in THF (20 mL) was added to a solution of Fe(HMDS)2·THF
(61.7 mg, 0.137 mmol) in THF (5 mL). An aliquot of the resulting
blue solution was taken and analyzed through NMR. 1H NMR (300.13
MHz, THF-d8, δ, ppm) 189.2 (1H, bs), 150.3 (1H, bs), 82.8 (1H, s),
60.1 (1H, s), 45.054 (1H, s), 28.0 (1H, s), 18.2 (18H, bs, HMDS),
11.5 (1H, s), 10.2 (1H, bs), 8.951 (1H, s), −2.3 (1H, s), −2.9 (1H, s),
−25.8 (1H, bs), −34.1 (1H, s).
General Procedures for the Catalytic Hydrosilylation of
Acetophenone. Inside the glovebox, compound 2 (0.25 or 0.5 mol
%) was dissolved in THF (6 mL) inside a vial or a Schlenk tube. The
corresponding silane (2.5 mmol) was then added to the solution,
followed by acetophenone (2 mmol) and the vial or Schlenk tube
sealed. The resulting mixture was stirred at room temperature during
the times indicated in Table 1 and then the reaction was quenched by
adding 1 M HCl (aqueous, 5 mL) while in an ice bath and stirred long
enough to ensure the complete hydrolysis of the silylether. The
mixture was then taken to ph 7 by adding 5 M NaOH (aqueous) and
the organic products extracted with ethyl acetate (3 × 15 mL). The
organic layers were dried using anhydrous Na2SO4, filtered, and
concentrated by rotatory evaporation. As an alternative method, inside
the glovebox, ligand 1 (0.55 mol %) and LiHMDS (0.50 mol %) were
dissolved together in THF (6 mL) inside a vial, proving the
deprotonation of ligand 1 by the formation of a deep blue solution.
Synthesis and Analytical Data for Ligand 1. To a solution of 1-
chloroanthraquinone (2.4 g, 9.69 mmol) in DMSO (30 mL) in a
Schlenk vessel was added 2-picolylamine (2.50 mL, 24.23 mmol), and
the mixture was stirred at 150 °C during 20 min. Then, the solution
was poured into cold distilled water (400 mL), and the resulting
precipitate was filtered out and washed with distilled water. The pure
product was obtained as a red powder after purification through silica
E
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
This solution was then poured inside another vial containing the
appropriate iron(II) precursor (0.25 mol %). To the resulting mixture,
the corresponding silane (2.5 mmol) was added, followed by
acetophenone (2 mmol) and the vial sealed. The rest of the procedure
was identical as that described above.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.6b00765.
■
S
X-ray Structure Determination. Prismatic crystals of compounds
1, 2, and 2b suitable for X-ray experiments were obtained and resin
epoxy coated and mounted on Bruker Axs APEX (1) and Stoe IPDS
2T (2 and 2b) diffractometers, respectively. Instruments were
equipped with graphite monochromated Mo Kα radiation (λ =
0.71073 Å), operating at 50 kV and a temperature of 293 K for 1 and
100 K for 2 and 2b. The cell parameters were determined and refined
by least-squares fit of all reflections collected. An empirical absorption
correction was applied. The data reduction was performed with the
APEX221 software and corrected for absorption using SADABS.22
Crystal structures were solved by direct methods using the SIR97
program23 and refined by full-matrix least-squares on F2 including all
reflections using anisotropic displacement parameters by means of the
WINGX crystallographic package.24 In all cases, the hydrogen atoms
were included with their calculated positions determined by molecular
geometry and refined riding on the corresponding bonded atom. Final
R(F), wR(F2), and goodness of fit agreement factors, details on the
data collection, and analysis can be found in Table S1. Cambridge
Crystallographic Data Centre numbers are 1493111 (1), 1507476 (2),
and 1514974 (2b). These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.
uk/data_request/cif.
Experimental procedures, spectral and crystallographic
data, and computational details (PDF)
Cartesian coordinates for all optimized structures (XYZ)
Crystallographic data for 1, 2, and 2b (CIF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail, cramer@umn.edu. Twitter, @ChemProfCramer
*E-mail: kuzui@staff.uni-marburg.de.
*E-mail: ifernan@ual.es.
■
ORCID
Christopher J. Cramer: 0000-0001-5048-1859
Ignacio Fernandez: 0000-0001-8355-580X
́
Present Address
́
́
̀
P.O.-B.: Instituto de Tecnologia Quimica, Universitat Politecn-
́
̀
ica de Valencia-Consejo Superior de Investigaciones Cientificas
(UPV-CSIC), Avda. de los Naranjos s/n, 46022 Valencia,
Spain.
Notes
Cyclic Voltammetry. For the electrochemical measurements, a
temperature-controlled microcell HC (RHD Instruments) was used.
The authors declare no competing financial interest.
The measurements were carried out at 25
0.1 °C in a three-
ACKNOWLEDGMENTS
This work was supported by the Junta de Andalucia (Spain)
under the project number P12-FQM-2668, the U.S. National
Science Foundation (CHE-1361595) and by Bruker Espanola
SA. A.R.-B. and P.O.-B. thank the University of Almeria and
MEC for a Ph.D. fellowship and a Ramon y Cajal contract
(RYC-2014-16620), respectively. R.L. is funded by the DFG
(LA 2830/3-2). I.K. would like to thank Prof. S. Dehnen for
■
electrode configuration with a polycrystalline Pt wires acting as a
pseudoreference and working electrode. A Pt crucible acted as a
container for the samples and as the counter electrode with
[nBu4N][PF6] (0.1M) as electrolyte. The electrochemical cell was
connected to a PGSTAT204 potentiostat/galvanostat (Metrohm
Autolab). In order to achieve the same conditions for all measure-
ments, the electrodes were freshly polished, carefully rinsed with
acetone and CH2Cl2, and finally dried in vacuum. For each sample, the
pure electrolyte solution was measured as a control experiment.
Ferrocene was used as internal standard, and the reported potentials
are referenced to the standard potential of [(η5-C5H5)2Fe]/[(η5-
C5H5)2Fe]+ (Fc/Fc+). Furthermore, each redox wave was examined
separately at different scan rates, showing a linear dependency of the
current versus the square root of the scan rate processes under
diffusion control. Table S4 summarizes the electrochemical data
obtained from the cyclic voltammetry measurements of ligand 1 and
complex 2.
Computational Details. All calculations were performed at the
density functional theory (DFT) level25 using the M06-L local
functional26 as implemented in Gaussian 09.27 Numerical integrations
were performed with an ultrafine grid. An automatic density-fitting set
generated by the Gaussian program was used to reduce computational
cost. Geometry optimizations were performed in the gas phase using
the def2-SVP basis sets.28 The natures of all stationary points were
confirmed by analytic computation of vibrational frequencies.
Transition state structures were verified to connect with the
corresponding reactants and products by following normal modes
associated with their imaginary frequencies. All frequencies below 50
cm−1 were replaced by 50 cm−1 when computing vibrational partition
functions.29 Final free energies in solution were computed by adding
gas-phase free energy contributions (298.15 K) to single-point
calculations in THF solvent using the SMD model30 and the def2-
TZVPP basis sets.28 For all species, a factor of RT·ln(24.46) was
included to take into account the 1 atm to 1 M standard-state free
energy change. Additional single-point calculations were carried using
the B3LYP-D3 hybrid functional (see Table S5).31,32
́
̃
́
́
generous support and Prof. B. Neumuller for X-ray measure-
̈
ments of 2 and 2b. We acknowledge the Minnesota
Supercomputing Institute (MSI) at the University of Minnesota
for providing resources that contributed to the research results
reported within this paper.
REFERENCES
■
(1) (a) Marciniec, B.; Gulinski, J.; Urbaniak, W.; Kornetka, Z. W.
Comprehensive Handbook on Hydrosilylation; Pergamon Press: Oxford,
U.K., 1992;. (b) Roy, A. K. Adv. Organomet. Chem. 2007, 55, 1−59.
(c) Marciniec, B., Ed. Hydrosilylation: A Comprehensive Review on
Recent Advances; Springer: Dordrecht, The Netherlands, 2009.
(2) (a) Marciniec, B.; Gulinsky, J.; Urbaniak, W.; Kornetka, Z. W. In
Comprehensive Handbook on Hydrosilylation; Marciniec, B., Ed.;
Pergamon: Oxford, 1992. (b) Marciniec, B., Ed. Hydrosilylation: A
Comprehensive Review on Recent Advances; Springer: New York, 2009.
For reviews of iron-catalyzed hydrosilylation: (c) Morris, R. H. Chem.
Soc. Rev. 2009, 38, 2282−2291. (d) Zhang, M.; Zhang, A. Appl.
Organomet. Chem. 2010, 24, 751−757. (e) Junge, K.; Schroder, K.;
Beller, M. Chem. Commun. 2011, 47, 4849−4859.
(3) Selected reviews: (a) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L.
Chem. Rev. 2004, 104, 6217−6254. (b) Plietker, B., Ed. Iron Catalysis
in Organic Chemistry; Wiley-VCH: Weinheim, Germany, 2008.
(c) Zhang, M.; Zhang, A. Appl. Organomet. Chem. 2010, 24, 751−
757. (d) Chakraborty, S.; Guan, H. Dalton Trans. 2010, 39, 7427−
7436. (e) Le Bailly, B. A. F.; Thomas, S. P. RSC Adv. 2011, 1, 1435−
1445. (f) Bullock, R. M. Catalysis without precious metals; Wiley-VCH:
Weinheim, 2010. (g) Bullock, R. M. Science 2013, 342, 1054−1055.
(h) Bauer, I.; Knolker, H.-J. Chem. Rev. 2015, 115, 3170−3387.
F
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(i) Hoyt, J. M.; Schmidt, V. A.; Tondreau, A. M.; Chirik, P. J. Science
2015, 349, 960−963. (j) Yu, R. P.; Hesk, D.; Rivera, N.; Pelczer, I.;
Chirik, P. J. Nature 2016, 529, 195−199.
(21) (a) APEX2, version 2008.1−0; Bruker AXS Inc.: Madison, WI,
2008. (b) SAINT. Data Reduction Software, V6.28A; Bruker AXS Inc.:
Madison, WI, 2001.
(22) Sheldrick, G. M. SADABS. Program for Empirical Absorption
Correction; University of Gottingen: Gottingen, Germany, 1996.
(23) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, G. L.;
Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 1999, 32, 115−119.
(4) Selected references on Fe-catalyzed hydrosilylations: (a) Ton-
dreau, A. M.; Darmon, J. M.; Wile, B. M.; Floyd, S. K.; Lobkovsky, E.;
Chirik, P. J. Organometallics 2009, 28, 3928−3940. (b) Inagaki, T.; Ito,
A.; Ito, J.; Nishiyama, H. Angew. Chem., Int. Ed. 2010, 49, 9384−9387.
(c) Kandepi, V. V. K. M.; Cardoso, J. M. S.; Peris, E.; Royo, B.
Organometallics 2010, 29, 2777−2782. (d) Yang, J.; Tilley, D. T.
Angew. Chem., Int. Ed. 2010, 49, 10186−10188. (e) Bhattacharya, P.;
Krause, J. A.; Guan, H. Organometallics 2011, 30, 4720−4729.
(f) Buitrago, E.; Tinnis, F.; Adolfsson, H. Adv. Synth. Catal. 2012,
354, 217−222. (g) 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. (h) Blom, B.; Enthaler, S.;
Inoue, S.; Irran, E.; Driess, M. J. Am. Chem. Soc. 2013, 135, 6703−
6713. (i) Zuo, Z.; Sun, H.; Wang, L.; Li, X. Dalton Trans. 2014, 43,
11716−11722. (j) Zuo, Z.; Zhang, L.; Leng, X.; Huang, Z. Chem.
Commun. 2015, 51, 5073−5076. (k) Wekesa, F. S.; Arias-Ugarte, R.;
Kong, L.; Sumner, Z.; McGovern, G. P.; Findlater, M. Organometallics
2015, 34, 5051−5056.
(24) (a) Sheldrick, G. M. Program for Crystal Structure Refinement;
University of Gottingen: Gottingen, Germany, 2014. (b) Farrugia, L. J.
̈
̈
J. Appl. Crystallogr. 1999, 32, 837−838.
(25) Cramer, C. J.; Truhlar, D. G. Phys. Chem. Chem. Phys. 2009, 11,
10757−10816.
(26) (a) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125, 194101.
(b) Zhao, Y.; Truhlar, D. G. Acc. Chem. Res. 2008, 41, 157−167.
(c) Zhao, Y.; Truhlar, D. G. Chem. Phys. Lett. 2011, 502, 1−13.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(5) Bleith, T.; Wadepohl, H.; Gade, L. H. J. Am. Chem. Soc. 2015,
137, 2456−2459.
(6) Nishiyama, H.; Furuta, A. Chem. Commun. 2007, 760−762.
(7) Wang, W.; Gu, P.; Wang, Y.; Wei, H. Organometallics 2014, 33,
847−857.
(8) (a) Gallego, D.; Inoue, S.; Blom, B.; Driess, M. Organometallics
2014, 33, 6885−6897. (b) Metsanen, T. T.; Gallego, D.; Szilvas
́
i, T.;
̈
Driess, M.; Oestreich, M. Chem. Sci. 2015, 6, 7143−7149.
(9) Bleith, T.; Gade, L. H. J. Am. Chem. Soc. 2016, 138, 4972−4983.
(10) (a) da Cunha, C. J.; Fielder, S. S.; Stynes, D. V.; Masui, H.;
Auburn, P. R.; Lever, A. B. P. Inorg. Chim. Acta 1996, 242, 293−302.
(b) Ochiai, K.; Mazaki, Y.; Nishikiori, S.; Kobayashi, K.; Hayashi, S. J.
Chem. Soc., Perkin Trans. 2 1996, 1139−1145. (c) Joshi, B. S.; Rho, T.;
Rinaldi, P. L.; Liu, W.; Wagler, T. A.; Newton, M. G.; Lee, D.;
Pelletier, S. W. J. Chem. Crystallogr. 1991, 27, 553−559.
(28) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3297−3305.
(29) Ribeiro, R. F.; Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J.
Phys. Chem. B 2011, 115, 14556.
(30) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B
2009, 113, 6378−6396.
(31) (a) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38,
3098−3100. (b) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens.
Matter Mater. Phys. 1988, 37, 785−789. (c) Stephens, P. J.; Devlin, F.
J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623.
(32) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
2010, 132, 154104.
(11) For related examples, see Ghosh, C.; Groy, T. L.; Bowman, A.
C.; Trovitch, R. J. Chem. Commun. 2016, 52, 4553−4556.
(12) Vergote, T.; Nahra, F.; Merschaert, A.; Riant, O.; Peeters, D.;
Leyssens, T. Organometallics 2014, 33, 1953−1963.
(13) Ghosh, C.; Mukhopadhyay, T. K.; Flores, M.; Groy, T. L.;
Trovitch, R. J. Inorg. Chem. 2015, 54, 10398−10406.
(14) Small spin contamination (<S2> of ca. 6.4 instead of 6.0) was
observed for [Fe] complexes, likely due to possible contribution of the
anthraquinonic ligand to higher spin states.
(15) [Ti]: (a) Bandini, M.; Bernardi, F.; Bottoni, A.; Cozzi, P. G.;
Miscione, G. P.; Umani-Ronchi, A. Eur. J. Org. Chem. 2003, 2003,
2972−2984. [Cu]: (b) Issenhuth, J.-T.; Notter, F.-P.; Dagorne, S.;
Dedieu, A.; Bellemin-Laponnaz, S. Eur. J. Inorg. Chem. 2010, 2010,
529−541. (c) Vergote, T.; Gathy, T.; Nahra, F.; Riant, O.; Peeters, D.;
Leyssens, T. Theor. Chem. Acc. 2012, 131, 1253.
(16) We have computed the species 4 and TS4−5 using the actual
silane, that is, (EtO)2MeSiH. The relative activation free energy is 27.3
kcal mol−1, which is essentially the same as the value reported for
(MeO)2MeSiH, 26.3 kcal mol−1.
(17) Kozuch, S. WIREs Comput. Mol. Sci. 2012, 2, 795−815. Our
value is estimated by considering both the catalytic cycle (hydro-
silylation) energies and the initial catalyst activation step at 298 K.
́
(18) (a) Gomez-Gallego, M.; Sierra, M. A. Chem. Rev. 2011, 111,
4857−4963. (b) Christensen, N. J.; Fristrup, P. Synlett 2015, 26, 508−
513.
(19) We note that H/D KIE might not be conclusive for
distinguishing the rate-limiting character of various elementary steps
as all of them involve hydride transfers. Instead, 13C-KIE studies for
the carbonyl carbon of acetophenone are in progress. We predict a
value of 1.025 for ketone reduction through TS6−7.
(20) Andersen, R. A.; Faegri, K., Jr.; Green, J. C.; Haaland, A.;
Lappert, M. F.; Leung, W.-P.; Rypdal, K. Inorg. Chem. 1988, 27, 1782−
1786.
G
DOI: 10.1021/acs.organomet.6b00765
Organometallics XXXX, XXX, XXX−XXX