A Highly Efficient Base-Metal Catalyst: Chemoselective Reduction of Imines to Amines Using An Abnormal-NHC-Fe(0) Complex

Article abstract of DOI:10.1021/acs.organomet.6b00478

A base-metal, Fe(0)-catalyzed hydrosilylation of imines to obtain amines is reported here which outperforms its noble-metal congeners with the highest TON of 17000. The catalyst, (aNHC)Fe(CO)₄, works under very mild conditions, with extremely low catalyst loading (down to 0.005 mol %), and exhibits excellent chemoselectivity. The facile nature of the imine reduction under mild conditions has been further demonstrated by reducing imines towards expensive commercial amines and biologically important N-alkylated sugars, which are difficult to achieve otherwise. A mechanistic pathway and the source of chemoselectivity for imine hydrosilylation have been proposed on the basis of the well-defined catalyst and isolable intermediates along the catalytic cycle.

Full text of DOI:10.1021/acs.organomet.6b00478

Article
pubs.acs.org/Organometallics
A Highly Efficient Base-Metal Catalyst: Chemoselective Reduction of
Imines to Amines Using An Abnormal-NHC−Fe(0) Complex
Mrinal Bhunia,† Pradip Kumar Hota,† Gonela Vijaykumar,† Debashis Adhikari,‡
and Swadhin K. Mandal*,†
†Department of Chemical Sciences, Indian Institute of Science Education and Research Kolkata, Mohanpur 741246, India
‡Department of Chemical Sciences, Indian Institute of Science Education and Research Mohali, SAS Nagar 140306, India
S
* Supporting Information
ABSTRACT: A base-metal, Fe(0)-catalyzed hydrosilylation of imines
to obtain amines is reported here which outperforms its noble-metal
congeners with the highest TON of 17000. The catalyst, (aNHC)Fe-
(CO)4, works under very mild conditions, with extremely low catalyst
loading (down to 0.005 mol %), and exhibits excellent chemoselectivity.
The facile nature of the imine reduction under mild conditions has been
further demonstrated by reducing imines towards expensive commercial
amines and biologically important N-alkylated sugars, which are difficult
to achieve otherwise. A mechanistic pathway and the source of
chemoselectivity for imine hydrosilylation have been proposed on the
basis of the well-defined catalyst and isolable intermediates along the
catalytic cycle.
mines are ubiquitously present in a wide variety of
chemicals, including natural products, agrochemicals,
dimer complex II by Brookhart7b and also the bowl-shaped
phosphine complex of rhodium III by Tsuji7c (Chart 1).
Consequently, an inexpensive and environmentally benign
transition-metal surrogate, replacing scarce, toxic, and expensive
transition metals, is a pressing need.8 In this direction, the use
of normal N-heterocyclic carbenes (nNHCs) has shown
promise in designing base-metal catalysts for imine reduction.9
Despite some interesting reports on iron-catalyzed hydro-
sylilation,9a,10 there is a significant lack of very active iron-based
catalysts which can in particular reduce ketimine at room
temperature.
Recently N-heterocyclic carbene and its close analogues
heralded a new era in base-metal catalysis and have enjoyed a
significant research focus.11 Crabtree and Bertrand have
pioneered the abnormal binding modes of N-heterocyclic
carbenes, which are proved to be even a better σ donors in
comparison to their classical normal NHC analogues.12 In the
past few years, we have established that the isolated abnormal
NHC (aNHC) is an exceptional ligand for designing useful
organic13 as well as organometallic catalysts14 for a broad
variety of organic transformations. Additionally, we were
motivated by the fact that, despite some sparse reports of
aNHC-based iron complexes,15 the catalytic activity of such
complexes thus far has not been thoroughly investigated. To
the best of our knowledge, a single report has been
documented by Darcel and co-workers involving a
cyclopentadienyliron(II)−normal NHC complex which works
A
pharmaceuticals, and biologically active compounds.1 The
importance of amines in biological processes is further
emphasized by their very frequent occurrence in different
alkaloids, amino acids, and nucleotides. Owing to the
widespread use of amines, the development of novel methods
toward their effective and economical preparation is a highly
desirable goal in catalysis research. For their preparation, the
catalytic hydrosilylation of nitriles, amides, imines, and
nitroarenes2 and reductive amination3 are among the very
few commonly practiced methods. Despite some sporadic
reports of transition metal catalyzed reduction of imines by
hydrogenation,4 hydrogen transfer,5 or hydrosilylation,6 a well-
performing catalyst to achieve this important transformation is
still missing and hence highly coveted. The few reported
catalysts for imine reduction are heavily dominated by rare,
expensive, and toxic late transition metal complexes (Chart 1).7
Some notable recent examples of catalytic hydrosilylation of
imines using late transition metals include the (μ-silane)-
diruthenium complex I by Kira,7a silylene-bridged iridium
Chart 1. Examples of Transition-Metal Catalysts Used for
Hydrosilylation of Imines
Received: June 13, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
for the reduction of ketimines at higher temperature (100
°C).9a We surmised that the superior σ-donor ability of aNHCs
may make the low-valent Fe(0) center sufficiently electron rich
so that oxidative addition of the silane can be easily achieved,
and this prompted us to use an (aNHC)Fe complex. We also
anticipated that the combination of electron-rich Fe(0) with a
strong σ donor aNHC ligand can work under very mild
conditions, which may allow the synthesis of an array of amine-
containing biologically important molecules that are usually
sensitive to harsh reaction conditions. Thus, in our pursuit for a
better iron catalyst, we report an elegant example of an iron(0)
complex, [(aNHC)Fe(CO)4] (1), which has a tremendous
potential for efficient hydrosilylation of imines at room
temperature with excellent chemoselectivity. The catalyst can
convert a plethora of imines into amines with an extremely low
catalyst loading of 0.05 mol %. This extremely low catalyst
loading leads to an impressive TON of 17000, effectively
outperforming all of the previous catalysts reported for imine
hydrosilylation.2f,9a More encouragingly, this is a relatively rare
entry where such a reactive system also demonstrate excellent
selectivity in reducing imine functionalities in the presence of
other commonly reducible carbonyl, cyano, and nitro motifs.
Such a combination of efficiency and selectivity in imine
reduction prompted us to reduce imines connected to sugar
moieties, resulting in biologically relevant N-alkylated sugars,
which are difficult to synthesize otherwise.
Figure 1. View of the molecular structure of 1. Ellipsoids are set at the
50% probability level; hydrogen atoms have been omitted for clarity.
Selected bond distances (Å) and angles (deg): Fe1−C5, 2.026(3);
Fe1−C42, 1.796(4); Fe1−C45, 1.770(4); Fe1−C44, 1.757(4); Fe1−
C43, 1.745(4); C42−Fe1−C5, 101.84(14); C45−Fe1−C5, 85.28(15);
C43−Fe1−C5, 169.71(16), C44−Fe1−C5, 84.82(15).
2.026(3) Å, comparable to that in [(NHC)Fe(CO)4],17a as
analyzed crystallographically.
After analyzing the structure by X-ray crystallography, we
used 1 as a catalyst for hydrosilylation reactions of imines.
Given the industrial importance of 1-aryl ethylamines, hydro-
silylation of N-benzylideneaniline as an aldimine partner was
chosen as a model substrate for this study with PhSiH3 as a
silylating agent. Gratifyingly, the reaction went smoothly at
room temperature to realize 95% conversion to the
corresponding amine in ethanol. Upon further optimization,
full conversion (>99%) was achieved in DMSO with a catalyst
loading of as low as 0.05 mol %. The role of the catalyst 1 in the
silylation of imines is eminent, as in a blank reaction no product
was detected by 1H NMR spectroscopy even at higher
temperature (Table S1, entry 23, in the Supporting
Information). We tested various silanes, including cheaper
silanes such as PMHS and (TMSO)2SiMeH. PMHS can be
used successfully, but it gave a good yield (around 70%) only
when it was used in excess (10 equiv), while the use of 2 equiv
of (TMSO)2SiMeH resulted in good to moderate yield (see
Table 1, entry 25 and Table S1). Among a variety of silanes
tested, PhSiH3 was the best, although other silanes provided
moderate conversion to silylated amines. Strikingly, the catalyst
can also perform with extremely low loading (0.005 mol %) at
room temperature, leading to a considerably high TON value of
17000, albeit with slightly lower yield (85%).
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The synthesis of 1 was
accomplished by the treatment of free aNHC with
commercially available diiron nonacarbonyl [Fe2(CO)9]) in a
2:1 stoichiometric ratio in toluene at room temperature
(Scheme 1). Analytically pure pale yellow crystals of 1 were
Scheme 1. Synthesis of Complex 1
obtained in 52% isolated yield from a concentrated toluene
solution at −21 °C. Compound 1 readily dissolves in toluene,
benzene, and THF and shows somewhat limited solubility in
hexane and pentane.
The nature of the complex has been probed by an array of
spectroscopic means (1H and 13C NMR and IR spectroscopy),
as well as by CHN analysis and single crystal X-ray diffraction
studies. The 13C NMR spectrum of 1 reveals a resonance at
218.2 ppm, assignable to the C-5 carbon bound to the iron
center which is expectedly shifted downfield from the
corresponding chemical shift of free aNHC.
Substrate Scope of Aldimines. With the optimized
reaction conditions in hand (Table S1, entry 16, in the
Supporting Information), we investigated the scope of the
reduction of various aldimines (2) to their corresponding
secondary amines (3) (Table 1). Under the standardized
conditions, the presence of an electron-donating group at the
para and ortho positions of the aniline moiety afforded high
yields (78−92%) (Table 1, entries 2−4). Under similar
conditions, benzylidene moieties containing both electron-
donating and -withdrawing substituents at the para position of
benzylidene offered a nearly quantitative yield of the
corresponding amines (Table 1, entries 5−10). Moreover, the
reduction was effective without any dehalogenations evidenced
from the substrates containing chloro, bromo, and iodo groups
To ascertain the bond connectivities, a medium sized crystal
was analyzed via single crystal X-ray diffraction (Figure 1).16
Compound 1 crystallizes in the triclinic space group P1 with a
̅
single molecule in the asymmetric unit. As expected, the X-ray
crystal structure of 1 exhibits an iron(0) trapped in a distorted-
trigonal-bipyramidal environment and bonded to the aNHC
and four carbonyl moieties. The Fe−Ccarbene bond distance is
B
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Hydrosilylation of Various Aldimines Catalyzed by
1
Table 1. continued
a
b
c
temperature, 12 h. Isolated yield. Complex 1 (0.2 mol %), 50 °C.
d
e
Conversion by 1H NMR spectroscopy. 63% and 66% 1H NMR
spectroscopic conversions with PMHS (10 equiv) and
(TMSO)2SiMeH (2 equiv), respectively.
(Table 1, entries 8−10). Substrates with highly electron
withdrawing groups, such as 4-cyano- and 4-nitrobenzylidene
derivatives, also afforded excellent isolated yields (85−95%)
with slightly high catalyst loading (0.2 mol %) and at elevated
temperature (50 °C) (Table 1, entries 11 and 12). Most
notably, the catalyst is considerably selective toward reduction
of imine in the presence of various carbonyl groups such as
ketone and ester at the 4-position of benzylidene, providing
excellent yields (85−92%) of the expected amine products
(Table 1, entries 13 and 14). In the presence of heavily electron
deficient groups such as 4-nitro and 4-cyano on the benzylidene
moiety, amine products were also furnished in excellent yield
(Table 1, entries 16 and 17). The catalyst can also perform very
well in the presence of an electron-rich substituent, as the
reaction with a 3,4,5-trimethoxy-substituted benzyl substrate
went smoothly to 96% isolated yield of N-(3,4,5-trimethox-
ybenzyl)-4-methoxyaniline (Table 1, entry 18). Gratifyingly,
this hydrosilylation methodology proved fruitful in synthesizing
biologically active amines in high yields, which are prohibitively
expensive from commercial sources (for entry 19, 50 mg costs
$670),18 arguably due to the dearth of elegant synthetic
methods (Table 1, entries 19 and 20). The catalyst’s
chemoselectivity was further demonstrated by the reduction
of aldimine in the presence of a CC bond, leading to the
corresponding unsaturated amine with good conversion,
retaining the nonreduced double bond (Table 1, entry 21).
Our exploration for substrate scope was further expanded to
heteroaromatic imines, which were reduced well to offer
potentially good bidentate ligands (Table 1, entry 22). The
catalytic efficiency was also surveyed in the presence of N-
cyclohexylamine and anthracen-9-ylmethylene moieties to give
their corresponding amines in moderate to very good yield
(59−95%) (Table 1, entries 23 and 24).
Since complex 1 proved itself as an extremely active catalyst
for aldimine reduction, we were also curious to test its catalytic
efficacy in hydrosilylation reactions of more challenging
substrates such as ketimines (4). Optimization of the best
conditions of ketimine reduction involves slightly higher
catalyst loading (2 mol %) in ethanol with a longer reaction
time, owing to the increased steric congestion and attenuated
electrophilicity of the ketimine (Table S2, entry 9, in the
Supporting Information). To check the catalytic strength of
complex 1, we choose 4-methyl-N-[1-(4-methylphenyl)-
ethylidene]aniline as a model substrate for ketimine hydro-
silylation reactions.
Substrate Scope of Ketimines. To further survey the
substrate scope for such reduction, ketimines were chosen
accordingly, bearing electron-rich as well as electron-deficient
groups at the 4-position of the ethylidene moiety. A
quantitative yield was obtained for such reduction reactions
to the corresponding secondary amines (Table 2, entries 1−4).
Under similar reaction conditions, an electron-donating group
at the para position of aniline, the ortho position of ethylidene,
and also an electron-withdrawing group at the 4-position of
acetophenone leads to the desired products in a promising
isolated yield (85−95%) (Table 2, entries 5−7). However,
a
Reaction conditions unless specified otherwise: aldimine (0.5 mmol),
PhSiH3 (1 mmol), and complex 1 (0.05 mol %), DMSO (1 mL), room
C
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2, hydrosilylations of 1,3,4,6-tetra-O-benzyl-2-(N-
benzylidene)-2-deoxy-β-D-glucose derivatives run smoothly
Table 2. Hydrosilylation of Various Ketimines Catalyzed by
1
a
Scheme 2. Hydrosilylation of Imines Bearing Sugar
a
Derivatives
a
Reaction conditions: imines (0.5 mmol), PhSiH3 (1 mmol), and
complex 1 (0.2 mol % for 6a−d and 1 mol % for 6e), DMSO (1 mL)
b
for 6a−d and EtOH (1 mL) for 6e, room temperature, 16 h. Isolated
c
1
yield. Conversion by H NMR spectroscopy.
with only 0.2 mol % of catalyst 1 loading. Several other sugar
derivatives of the imine can also be effectively reduced to obtain
the corresponding N-alkylated sugars in low catalyst loadings,
proving this reduction to be very mild and tolerant to sensitive
protections. Interestingly, when a sugar-containing heterocyclic
ring in the imine moiety was subjected to hydrosilylation, a 92%
yield of the corresponding amine product 7d was obtained at
room temperature. However, our catalytic protocol did not
work for unprotected sugar substrates containing free OH
functionalities. We note in passing that the imine reduction
methodology in protected sugars to prepare valuable N-
alkylated sugars closely resembles the in situ reduction of
imines during alcohol to amine direct coupling by borrowing
the hydrogen strategy developed by Feringa and Barta.22 In
their presented catalytic cycle, an alcohol is dehydrogenated to
the respective carbonyl compound, which upon reaction to
amine forms an imine, before being reduced by the trapped
hydrogen in the presence of an iron catalyst. Beller’s primary
amine synthesis from secondary alcohols using ammonia with a
ruthenium catalyst23 also indicates the intermediacy of an
imine.
Chemoselectivity of the Reduction Process. Our initial
findings presented in Tables 1 and 2 demonstrated that the
catalyst 1 not only is highly reactive but also can spare a
number of functional groups during the catalytic process,
leading to excellent chemoselectivity. The chemoselectivity of
the catalyst 1 is evident from the reduction of aldimines
possessing a nitrile, nitro, ketone, ester, acid group and an
unsaturated double bond (Table 1, entries 10−13, 15, 16, 19,
and 20). To our delight, in all cases these aldimines were
selectively reduced without affecting the presence of other
highly reducible groups. The only detrimental effect of
introducing an electron-withdrawing group in the aldimine
was the decelerated reaction rate. The excellent chemo-
selectivity in the reduction of imines was further scrutinized
by performing the catalytic reaction in the presence of an 1:1
mixture of an imine and the corresponding aldehyde or ketone,
a
Reaction conditions: ketimine (0.5 mmol), PhSiH3 (1 mmol),
complex 1 (2 mol %), ethanol (1 mL), room temperature, 24 h.
b
c
1
Isolated yield. Conversion by H NMR spectroscopy.
further tests of functional group tolerance in the presence of
reducible groups such as −NO2 and −CN resulted in poor
conversion (∼15%) to the corresponding amine products.
From these observations, it can possibly be inferred that steric
hindrance on the ethylidene moiety does not play a large role in
these reactions. It is noteworthy that the more challenging
substrates having an alkyl group in the ethylidene as well as in
amine moieties led to the expected reduction products in an
excellent isolated yield (85−93%, Table 2, entries 8−10).
Substrate Scope of Imines Bearing Sugar Moieties.
Our success in reducing multifarious aldimines and ketimines
by 1 under mild conditions prompted us to test the catalyst
further for reduction reactions in the presence of sugars. Sugar
derivatives are important precursors for biologically relevant
compounds such as glycosidase inhibitors, GlcN-6-P synthase
inhibitors, etc.19 It has been shown previously that strong
GlcN-6-P synthase inhibitory properties and good antifungal
activities were achieved by the introduction of N-alkyl
derivatives to 2-amino-2-deoxy-D-glucose.20 In the synthesis
of this N-alkylated sugar derivative, direct alkylation of 2-
amino-2-deoxy-D-glucose has been proven to be very difficult
and crumbersome.21 To overcome this problem, we have
introduced hydrosilylation for the preparation of N-alkyl
derivatives of 2-amino-2-deoxy-D-glucose. As presented in
D
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
which resulted in a clean reduction of only imines, leaving the
aldehyde or ketone intact (Scheme 3). This is in contrast to the
characterization for the elusive 9 prompted us to scrutinize it
computationally. The computationally optimized (B3LYP/
LanL2DZ(Fe)/6-31G*(H,C,N,O,Si) level of theory)24 struc-
ture (slightly truncated model) clearly exposes the hydride
group with an Fe−H distance of 1.53 Å (Figure 2), closely
Scheme 3. Chemoselective Reduction of Imines into Amines
a
in the Presence of an Aldehyde or Ketone
a
The corresponding alcohol was not obtained from aldehyde or
ketone reduction in absence of aldimine or ketimine.
fact that a less reactive catalyst, (IMS)Fe(CO)4, reduces
benzaldehyde at room temperature, but we attribute this fact to
the remarkable selectivity of our catalyst. To further understand
the source of such chemoselectivity, we resorted to high-level
DFT calculations (vide infra).
Figure 2. Computationally optimized structure of intermediate 9 and
the TS for hydride insertion to the imine. All hydrogens except for the
iron hydride and the silyl hydrogens have been omitted for clarity.
Mechanistic Understanding. The remarkable activity of 1
as a catalyst for the efficient hydrosilylation of imines inspired
us to uncover the mechanistic details of the above reaction. To
develop a clear picture for some of the intermediates en route
to the product formation, we investigated the detail via
stoichiometric reactions. At first, the catalyst 1 requires
activation by extruding one of the strongly π acidic CO ligands,
where the vacant site may facilitate the oxidative addition of the
silane. Upon CO loss, activated 8 is prone to oxidatively add
PhSiH3 to generate an iron hydride species along with a
coordinated silyl ligand (Scheme 4). The resulting hydride
resembling the Fe−H distances (1.43 and 1.46 Å) in
(IMes)Fe(H)(SiPh3)(CO)3 and (PMe2Ph)Fe(H)(SiPh3)-
(CO)3 as crystallographically characterized by Darcel.17a
Additionally, the theoretically derived value of the NMR
shielding tensor, −5.43 ppm, by gauge-independent atomic
orbital (GIAO) calculations for the iron hydride correlates well
with the experimentally observed value.25 The intermediate 9
was further characterized by 29Si NMR spectroscopy, displaying
a resonance at −6.07 ppm, as well as carbonyl stretches at 2003,
1963, and 1893 cm−1 by IR spectroscopy (Figures S37 and S38
in the Supporting Information). These CO stretches are
observed at higher frequencies γCO : 2003, 1963, and 1893 cm−1
in comparison to the stretches from 1 γCO : 1998, 1912, and
1871 cm−1, corroborating that the iron center has reached the
+2 oxidation state upon oxidative addition. Ideally, the ease of
oxidative addition of Fe(0) to Fe(II) should be facilitated by an
electron-rich metal center. To check this possibility, we
compared the head to head catalytic activity of an Fe(0)
complex bearing a normal NHC and compared it with that of
the present abnormal NHC bound Fe(0) catalyst in the
hydrosilylation of imines. Accordingly, the nNHC−Fe(0)
complex [(IPr)Fe(CO)4]17c was employed for the hydro-
silylation of two aldimines, 4-chlorobenzylidene-4-methylani-
line and 4-methylbenzylidene-4-methylaniline, furnishing 27%
and 43% yields of the corresponding amines, respectively,
under reaction conditions identical with those adopted for
catalyst 1. An analogous probe for ketimine hydrosilylation
using 4-methyl-N-(1-p-tolylethylidene)aniline as a test substrate
resulted in only a 35% yield of the corresponding amine. Both
the aldimine and ketimine reductions described above prove
irrefutably that our catalyst 1 performs significantly better
(yields 94%, 95%, and 89%, respectively; see Table 1, entries 5
and 10, and Table 2, entry 2) than its classical normal carbene
congener. We attribute this effect to the better σ-donor ability
of aNHC in comparison to nNHC, which will make the iron(0)
center more electron rich to facilitate oxidative addition of the
silane. Indeed, close analysis of the Tolman electronic
parameter26 clearly affirms our hypothesis. In the case of 1,
the γCO(A1) mode was observed at 1998 cm1, which is
Scheme 4. Plausible Catalytic Cycle for the Hydrosilylation
of Imines Catalyzed by Complex 1
intermediate species 9 is irrevocably identified from its
1
diagnostic H NMR chemical shift at −9.22 ppm (3JH,H = 1
Hz) along with the silyl hydrogen resonances at 5.36 ppm. A
very similar upfield shift for the Fe hydride (−8.74 ppm)17a that
has been reported earlier for an analogous complex comprising
a classical carbene supports our assertion further. The low-spin
Fe hydride complex 9 is unstable, denying us from crystallizing
it despite multiple attempts. The lack of crystallographic
E
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
considerably lower than 2035 cm−1, seen for [(IPr)Fe(CO)4].
This lower stretching frequency of the specific mode can clearly
be attributed to the enhanced π back-donation to the low-lying
π* orbital of CO, in the case of 1, owing to the greater σ
donation of the aNHC. Moreover, theoretically calculated CO
stretching frequencies for the γCO band of 1 (2003 cm−1) are
fully consistent with the above ([(IPr)Fe(CO)4], 2015 cm−1)27
trend.
In the intermediate 10 (Scheme 4), the imine substrate binds
to the iron center upon further creation of a vacant site by
removing CO from 9. The excellent chemoselectivity in imine
reduction elicited by 1 (vide supra) most likely lies in the
preferential binding of an imine functionality to the iron center.
As probed computationally, binding of the imine group in 4-
acetyl-N-(1-phenylbenzylidene)aniline is favored over the
carbonyl binding by 5.69 kcal mol−1 (see Figure 3). Exactly
gathering convincing data to delineate the plausible mechanism
for imine reduction.
CONCLUSIONS
■
In conclusion, we have developed an efficient aNHC based
iron(0) catalyst, which can selectively hydrosilylate various
aldimines and ketimines under remarkably low catalyst loadings
at room temperature with significantly high turnover numbers
(up to 17000). Since the catalysis occurs under very mild
conditions, the reduction can encompass a wide range of imine
substrates with considerable functional group tolerance, leading
to excellent chemoselectivity. Moreover, this catalyst exhibits
superior performance for the hydrosilylation of highly function-
alized imine substrates derived from sugars, to provide
straightforward access to their N-alkylated derivatives, which
have not been obtained earlier under mild conditions. In
addition, the present catalyst delivered excellent yield to
commercially available and expensive hydroxyl-functionalized
amines, which are difficult to achieve otherwise. A combined
experimental and theoretical investigation discloses that the
imine hydrosilylation reaction proceeds through the formation
of iron hydride complex 9 followed by hydride migration to
insert imines. The excellent σ-donating ability of the aNHC
ligand helps the iron center to be sufficiently electron rich so
that it can outperform its classical regular NHC congeners in
hydrosilylation reactions. Encouragingly, the observed catalytic
activity for the base-metal iron complex is even superior to that
of reported noble-metal catalysts used for hydrosilylation of
imines.
Figure 3. Computationally optimized structures of two binding modes
of imine (left) vs carbonyl to FeII (right). Imine binding is favored by
5.69 kcal mol−1 over the carbonyl binding mode. All hydrogens except
for the iron hydride and the silyl hydrogens have been omitted for
clarity.
EXPERIMENTAL SECTION
■
Methods. The ligand aNHC was prepared following the literature
method,12d and the nNHC iron(0) complex [(IPr)Fe(CO)4] was
prepared according to the literature procedure.17c The HRMS data
were obtained using a Finnigan MAT 8230 instrument. Elemental
analyses were carried out using a PerkinElmer 2400 CHN analyzer,
and samples were prepared by keeping them under reduced pressure
(10−2 mbar) overnight. The melting point was measured in a sealed
the same trend is observed for other aldimine substrates
containing aldehyde and ester functionalities, where binding of
imine is distinctly preferred over aldehyde or ester binding by
4.64 and 16.94 kcal mol−1, respectively, providing further
support to this hypothesis. The imine is subsequently inserted
by a hydride migration. Computationally, we were able to
locate the transition state (TS) for imine insertion, which poses
a free energy barrier of 21.6 kcal mol−1, with respect to the
imine-bound intermediate 9. After the insertion, reductive
elimination ensues, traverses through a barrier of 13.5 kcal
mol−1 to liberate the silylamine product, and reverts back to the
catalyst at the Fe(0) stage. Overall, our proposed mechanism
follows the classic Chalk−Harrod pathway of hydrosylilation,28
where oxidative addition of the silane and hydride migration to
insert the imine are the two key steps. Notably, the pivotal role
of the Fe hydride species 9 in the catalytic cycle is
unambiguously established by its stoichiometric reaction with
N-(4-methoxybenzylidene)-4-methylaniline in DMSO-d6 at
slightly elevated temperature, in which the hydride resonance
vanishes completely to form 11, as traced by 1H NMR
spectroscopy. In our case, the alternative pathway of silyl
migration to incorporate imine (as opposed to hydride
migration) can be easily refuted, as the benzylic protons,
obtainable through hydride migration in the reduced imine, are
clearly displayed at 4.13 ppm in the 1H NMR spectrum (see the
Supporting Information). Of note, the well-defined catalyst
molecule 1 and isolable intermediate 9 inarguably help in
glass tube on a Buchi B-540 melting point apparatus. Crystallographic
̈
data for the structural analysis of 1 have been deposited at the
Cambridge Crystallographic Data Center (CCDC No. 1451782).
These data can be obtained free of charge from the Cambridge
Crystallographic Data Center. NMR spectra were recorded on a JEOL
ECS 400 MHz spectrometer and on a Bruker Avance III 500 MHz
spectrometer.
Synthesis of (aNHC)Fe(CO)4 (1). In a glovebox, an oven-dried 50
mL Schlenk flask was charged with Fe2(CO)9 (200 mg, 0.55 mmol)
and aNHC (541 mg, 1 mmol), and then dry toluene (25 mL) was
added via cannula at 25 °C under an argon atmosphere. The reaction
mixture was stirred overnight, and during the course of the reaction,
the color changed to yellowish red. Then the reaction mixture was
filtered through a Celite pad and concentrated further to ca. 8 mL.
Storage of this reaction mixture at −20 °C for 2−3 days afforded
yellow crystals (370 mg, 0.52 mmol, 52%). 1H NMR (C6D6, 500 MHz,
298 K): δ 7.85 (d, 2H, J = 7.5 Hz), 7.36 (t, 1H, J = 7.5 Hz), 7.15 (s,
2H), 7.07 (t, 2H, J = 8 Hz), 6.95- 6.91 (m, 2H), 6.81 (d, 2H, J = 7.5
Hz), 6.76 (d, 2H, J = 8 Hz), 6.57−6.54 (m, 1H), 6.51−6.48 (m, 2H),
2.99 (sept, 2H, J = 6.5 Hz), 2.93 (sept, 2H, J = 6.5 Hz), 1.60 (d, 6H, J
= 6.5 Hz), 0.99 (d, 6H, J = 7 Hz), 0.84 (d, 6H, J = 7 Hz), 0.69 (d, 6H,
J = 6.5 Hz) ppm. 13C NMR (C6D6, 125 MHz, 298 K): δ 218.2, 160.4,
146.6, 145.3, 144.8, 140.9, 136.6, 133.1, 132.3, 131.6, 131.1, 129.9,
129.6, 128.5, 128.4, 127.9, 127.5, 125.5, 125.2, 124.8, 29.2, 28.8, 24.4,
24.3, 23.8, 23.5 ppm. Anal. Calcd for C43H44FeN2O4: C, 72.88; H,
6.26; N, 3.95. Found: C, 72.72; H, 7.01; N, 3.90. IR (film; νCO, cm−1):
1998, 1912, 1871.
F
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
General Procedure for Catalytic Hydrosilylation of Aldi-
mines. A Schlenk tube was charged with the catalyst 1 (0.00025
mmol, 0.177 mg (prepared from a standard solution of 1 in DMSO),
0.05 mol %), aldimine reagents (0.5 mmol), PhSiH3 (1.0 mmol, 123.4
μL), and DMSO (1.0 mL) with a magnetic stirring bar in air. The
reaction mixture was stirred at room temperature for 12 h. After the
completion of the reaction, the reaction mixture was worked up with
ether and water. The organic solvent was evaporated under vacuum,
and the desired product was purified by column chromatography.
General Procedure for Catalytic Hydrosilylation of Keti-
mines. A Schlenk tube was charged with the catalyst 1 (0.01 mmol,
7.09 mg, 2.0 mol %), ketimine reagents (0.5 mmol), PhSiH3 (1.0
mmol, 123.4 μL), and HPLC-grade ethanol (1.0 mL) with a magnetic
stirring bar in air. The reaction mixture was stirred at room
temperature for 24 h. After completion of the reaction, the reaction
mixture was worked up with ether and water, the organic solvent was
evaporated under vacuum, and the desired product was purified by
column chromatography.
Jr. Org. Lett. 2005, 7, 5087−5090. (i) Tormo, J.; Hays, D. S.; Fu, G. C.
J. Org. Chem. 1998, 63, 5296−5297.
(3) (a) Pagnoux-Ozherelyeva, A.; Pannetier, N.; Mbaye, M. D.;
Gaillard, S.; Renaud, J.-L. Angew. Chem., Int. Ed. 2012, 51, 4976−4980.
(b) Steinhuebel, D.; Sun, Y.; Matsumura, K.; Sayo, N.; Saito, T. J. Am.
Chem. Soc. 2009, 131, 11316−11317.
(4) (a) Spindler, F.; Blaser, H.-U. In Transition Metals for Organic
Synthesis, 2nd ed.; Beller, M., Bolm, C., Eds.; Wiley-VCH: Weinheim,
Germany, 2004; Vol. 2, pp 113−123. (b) Blaser, H.-U.; Spindler, F. In
Handbook of Homogeneous Hydrogenation; Vries, J. G. D., Elsevier, C. J.,
Eds.; Wiley-VCH: Weinheim, Germany, 2007; Vol. 3, p 1193.
(c) Blaser, H.-U.; Malan, C.; Pugin, B.; Spinder, F.; Steiner, H.;
Studer, M. Adv. Synth. Catal. 2003, 345, 103−151. (d) Tang, W.;
Zhang, X. Chem. Rev. 2003, 103, 3029−3069. (e) Fabrello, A.;
Bachelier, A.; Urrutigoïty, M.; Kalck, P. Coord. Chem. Rev. 2010, 254,
273−287.
(5) (a) Gladiali, S.; Alberico, E. In Transition Metals for Organic
Synthesis, 2nd ed.; Beller, M., Bolm, C., Eds.; Wiley-VCH: Weinheim,
Germany, 2004; Vol. 2, pp 145−166. (b) Clapham, S. E.; Hadzovic, A.;
Morris, R. H. Coord. Chem. Rev. 2004, 248, 2201−2237. (c) Zuo, W.;
Lough, A. J.; Li, Y. F.; Morris, R. H. Science 2013, 342, 1080−1083.
(d) Zuo, W. W.; Morris, R. H. Nat. Protoc. 2015, 10, 241−257.
(e) Arita, S.; Koike, T.; Kayaki, Y.; Ikariya, T. Organometallics 2008, 27,
2795−2802.
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.6b00478.
́
(6) (a) Bezier, D.; Sortais, J.-B.; Darcel, C. Adv. Synth. Catal. 2013,
Experimental procedures, spectroscopic data, scanned
spectra, details of calculated structures, and X-ray
crystallographic data for 1 (PDF)
Cartesian coordinates of calculated structures(XYZ)
X-ray crystallographic data for 1 (CIF)
355, 19−33. (b) Li, B.; Sortais, J.-B.; Darcel, C.; Dixneuf, P. H.
ChemSusChem 2012, 5, 396−399. (c) Brunner, H.; Becker, R. Angew.
Chem., Int. Ed. Engl. 1984, 23, 222−223. (d) Li, B.; Bheeter, C. B.;
Darcel, C.; Dixneuf, P. H. ACS Catal. 2011, 1, 1221−1224.
(7) (a) Hashimoto, H.; Aratani, I.; Kabuto, C.; Kira, M. Organo-
metallics 2003, 22, 2199−2201. (b) Cheng, C.; Brookhart, M. J. Am.
Chem. Soc. 2012, 134, 11304−11307. (c) Niyomura, O.; Iwasawa, T.;
Sawada, N.; Tokunaga, M.; Obora, Y.; Tsuji, Y. Organometallics 2005,
24, 3468−3475.
AUTHOR INFORMATION
Corresponding Author
*E-mail for S.K.M.: swadhin.mandal@iiserkol.ac.in.
■
(8) (a) Lipshutz, B. H.; Shimizu, H. Angew. Chem., Int. Ed. 2004, 43,
2228−2230. (b) Wang, L.; Sun, H.; Li, X. Eur. J. Inorg. Chem. 2015,
2015, 2732−2743.
Notes
The authors declare no competing financial interest.
(9) (a) Castro, L. C. M.; Sortais, J.-B.; Darcel, C. Chem. Commun.
2012, 48, 151−153. (b) Bheeter, L. P.; Henrion, M.; Chetcuti, M. J.;
Darcel, C.; Ritleng, V.; Sortais, J.-B. Catal. Sci. Technol. 2013, 3, 3111−
3116.
ACKNOWLEDGMENTS
■
We thank the SERB (DST) of India (Grant No. SR/S1/IC-25/
2012) for financial support. M.B. and G.V. thank the UGC,
New Delhi, India, for research fellowships. P.K.H. thanks the
IISER Kolkata for a research fellowship. M.B. thanks Dr. Arup
Mukherjee and Dr. Arunabha Thakur for constructive
suggestions during this work. D.A. acknowledges the computa-
tional facility at Indiana University.
(10) (a) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc.
2004, 126, 13794−13807. (b) Archer, A. M.; Bouwkamp, M. W.;
Cortez, M.-P.; Lobkovsky, E.; Chirik, P. J. Organometallics 2006, 25,
4269−4278. (c) 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. (d) Kamata, K.; Suzuki, A.; Nakai, Y.; Nakazawa, H.
Organometallics 2012, 31, 3825−3828. (e) Belger, C.; Plietker, B.
Chem. Commun. 2012, 48, 5419−5421. (f) Kamitani, M.; Itazaki, M.;
Tamiya, C.; Nakazawa, H. J. Am. Chem. Soc. 2012, 134, 11932−11935.
(g) Li, Y.; Yu, S.; Wu, X.; Xiao, J.; Shen, W.; Dong, Z.; Gao, J. J. Am.
Chem. Soc. 2014, 136, 4031−4039. (h) Naumov, R. N.; Itazaki, M.;
Kamitani, M.; Nakazawa, H. J. Am. Chem. Soc. 2012, 134, 804−807.
(11) (a) N-Heterocyclic Carbenes in Transition Metal Catalysis;
Glorius, F., Ed.; Springer: Berlin, 2007; Topics in Organometallic
REFERENCES
■
(1) (a) Lawrence, S. A. Amines: Synthesis Properties and Applications;
Cambridge University Press: Cambridge, U.K., 2004. (b) Hayes, K. S.
Appl. Catal., A 2001, 221, 187−195. (c) Mueller, T. E.; Hultzsch, K.
C.; Yus, M.; Foubelo, F.; Tada, M. Chem. Rev. 2008, 108, 3795−3892.
(d) Hannedouche, J.; Schulz, E. Chem. - Eur. J. 2013, 19, 4972−4985.
(e) Mukherjee, A.; Srimani, D.; Chakraborty, S.; Ben-David, Y.;
Milstein, D. J. Am. Chem. Soc. 2015, 137, 8888−8891. (f) Balaraman,
E.; Srimani, D.; Diskin-Posner, Y.; Milstein, D. Catal. Lett. 2015, 145,
139−144.
(2) (a) Calas, R. Pure Appl. Chem. 1966, 13, 61−80. (b) Chalk, A. J. J.
Organomet. Chem. 1970, 21, 207−213. (c) Zhou, S.; Junge, K.; Addis,
D.; Das, S.; Beller, M. Angew. Chem., Int. Ed. 2009, 48, 9507−9510.
(d) Sunada, Y.; Kawakami, H.; Imaoka, T.; Motoyama, Y.; Nagashima,
H. Angew. Chem., Int. Ed. 2009, 48, 9511−9514. (e) Selvakumar, K.;
Rangareddy, K.; Harrod, J. F. Can. J. Chem. 2004, 82, 1244−1248.
(f) Corre, Y.; Iali, W.; Hamdaoui, M.; Trivelli, X.; Djukic, J.-P.;
Agbossou-Niedercorn, F.; Michon, C. Catal. Sci. Technol. 2015, 5,
1452−1458. (g) Junge, K.; Wendt, B.; Shaikh, N.; Beller, M. Chem.
Commun. 2010, 46, 1769−1771. (h) Rahaim, R. J., Jr.; Maleczka, R. E.,
Chemistry 21. (b) Diez-Gonzal
Rev. 2009, 109, 3612−3676. (c) Riener, K.; Haslinger, S.; Raba, A.;
Hogerl, M. P.; Cokoja, M.; Herrmann, W. A.; Kuhn, F. E. Chem. Rev.
́
ez, S.; Marion, N.; Nolan, S. P. Chem.
̈
̈
2014, 114, 5215−5272. (d) Mo, Z.; Liu, Y.; Deng, L. Angew. Chem.,
Int. Ed. 2013, 52, 10845−10849.
(12) (a) Grundemann, S.; Kovacevic, A.; Albrecht, M.; Faller, J. W.;
̈
Crabtree, R. H. Chem. Commun. 2001, 2274−227. (b) Albrecht, M.
Chem. Commun. 2008, 3601−3610. (c) Schuster, O.; Yang, L.;
Raubenheimer, H. G.; Albrecht, M. Chem. Rev. 2009, 109, 3445−3478.
(d) Aldeco-Perez, E.; Rosenthal, A. J.; Donnadieu, B.; Parameswaran,
P.; Frenking, G.; Bertrand, G. Science 2009, 326, 556−559.
(13) (a) Sen, T. K.; Sau, S. C.; Mukherjee, A.; Modak, A.; Mandal, S.
K.; Koley, D. Chem. Commun. 2011, 47, 11972−11974.
G
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(14) (a) Sau, S. C.; Santra, S.; Sen, T. K.; Mandal, S. K.; Koley, D.
Chem. Commun. 2012, 48, 555−557. (b) Sau, S. C.; Raha Roy, S.; Sen,
T. K.; Mullangi, D.; Mandal, S. K. Adv. Synth. Catal. 2013, 355, 2982−
2991. (c) Sau, S. C.; Raha Roy, S.; Mandal, S. K. Chem. - Asian J. 2014,
9, 2806−2813. (d) Raha Roy, S.; Sau, S. C.; Mandal, S. K. J. Org. Chem.
2014, 79, 9150−9160. (e) Hota, P. K.; Kumar, G. V.; Pariyar, A.; Sau,
S. C.; Sen, T. K.; Mandal, S. K. Adv. Synth. Catal. 2015, 357, 3162−
3170.
(15) (a) Day, B. M.; Pugh, T.; Hendriks, D.; Guerra, C. F.; Evans, D.
J.; Bickelhaupt, F. M.; Layfield, R. A. J. Am. Chem. Soc. 2013, 135,
13338−13341. (b) Lavallo, V.; El-Batta, A.; Bertrand, G.; Grubbs, R.
H. Angew. Chem., Int. Ed. 2011, 50, 268−271. (c) Danopoulos, A. A.;
Tsoureas, N.; Wright, J. A.; Light, M. E. Organometallics 2004, 23,
166−168.
(16) Crystal data for 1: C43H44FeN2O4, Mr = 708.65, triclinic, space
group P1, a = 10.1393(8) Å, b = 10.7878(10) Å, c = 18.1314(15) Å, α
̅
= 77.341(8)°, β = 80.516(7)°, γ = 79.342(7)°, V = 1885.6(3) nm3, Z =
4, ρcalc = 1.248 Mg m−3, μ(Mo Kα) = 0.443 mm−1, T = 100(2) K, θ
range for data collection 1.96−25.03°, R1 = 0.0529 (I > 2σ(I)), wR2 =
0.1225 (all data).
(17) (a) Li, H.; Misal Castro, L. C.; Zheng, J.; Roisnel, T.; Dorcet, V.;
Sortais, J.-B.; Darcel, C. Angew. Chem., Int. Ed. 2013, 52, 8045−8049.
(b) Huttner, G.; Gartzke, W. Chem. Ber. 1972, 105, 2714−2725.
(c) Warratz, S.; Postigo, L.; Royo, B. Organometallics 2013, 32, 893−
897.
(18) http://www.sigmaaldrich.com/catalog/product/aldrich/
s859613?lang=en&region=IN; accessed on February 6, 2016.
(19) (a) Lillelund, V. H.; Jensen, H. H.; Liang, X.; Bols, M. Chem.
Rev. 2002, 102, 515−553. (b) Maxwell, V. L.; Evinson, E. L.;
Emmerson, D. P. G.; Jenkins, P. R. Org. Biomol. Chem. 2006, 4, 2724−
2732.
(20) Liberek, B.; Melcer, A.; Osuch, A.; Wakiec,
́
R.; Milewski, S.;
Wisniewski, A. Carbohydr. Res. 2005, 340, 1876−1884.
́
(21) (a) Gibbs, C. F.; Hough, L.; Richardson, A. C. Carbohydr. Res.
1965, 1, 290−296. (b) Gigg, R.; Conant, R. Carbohydr. Res. 1982, 100,
C5−C9. (c) Borch, R. F.; Bernstein, M. D.; Durst, H. D. J. Am. Chem.
Soc. 1971, 93, 2897−2904. (d) Salvatore, R. N.; Hwan, Y. Ch.; Woon,
J. K. Tetrahedron 2001, 57, 7785−7811.
(22) (a) Bahn, S.; Imm, S.; Neubert, L.; Zhang, M.; Neumann, H.;
Beller, M. ChemCatChem 2011, 3, 1853−1864. (b) Yan, T.; Feringa, B.
L.; Barta, K. Nat. Commun. 2014, 5, 5602−5608. (c) Yan, T.; Feringa,
B. L.; Barta, K. ACS Catal. 2016, 6, 381−388.
(23) Imm, S.; Bahn, S.; Neubert, L.; Neumann, H.; Beller, M. Angew.
̈
Chem., Int. Ed. 2010, 49, 8126−8129.
(24) (a) Becke, A. D. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38,
3098−3100. (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(25) GIAO shielding constants σ were converted to chemical shifts
using δ = σTMS − σ.
(26) (a) Gusev, D. G. Organometallics 2009, 28, 6458−6461.
(b) Gusev, D. G. Organometallics 2009, 28, 763−770.
(27) Computed frequencies were scaled with a multiplication factor
of 0.95.
(28) Chalk, A. J.; Harrod, J. F. J. Am. Chem. Soc. 1965, 87, 16−21.
H
DOI: 10.1021/acs.organomet.6b00478
Organometallics XXXX, XXX, XXX−XXX