Chemoselective Reduction of Imines Catalyzed by Ruthenium(II) Half-Sandwich Complexes: A Mechanistic Study

Article abstract of DOI:10.1002/ejic.201900455

Ruthenium half-sandwich complexes ligated to chiral 2-oxazolidinethiones or 2-thiozolidinethiones in the reduction of N-benzylideneaniline using silyl hydrides as reductants has been examined. The chemoselective reduction of imines takes place under mild conditions to afford the corresponding amines in nearly quantitative yield. Mechanistic studies indicate that dissociation of the ancillary ligands generate the active catalyst in all the complexes studied, which is the same species generated by [Ru(p-cymene)(Cl)₂]₂ under the reaction conditions. This results in the formation of a single catalytic species irrespective of the starting half-sandwich complex. Detailed mechanistic studies involving trapping of intermediates, in situ studies using mass spectrometry and NMR spectroscopy were carried out using the active catalyst generated by [Ru(p-cymene)(Cl)₂]₂. The mechanism of the reaction is dependent on the number of the hydrogen atoms in the reducing silane. The reaction proceeds via Gade-Hoffman pathway or Zheng-Chan pathway when a dihydro or trihydrosilane is the reductant. However, the use of a monohydrosilane, leads to longer reaction times presumably due to a change in the reaction pathway.

Full text of DOI:10.1002/ejic.201900455

A Journal of
Accepted Article
Title: Chemoselective reduction of imines catalyzed by Ru(II) half-
sandwich complexes: a mechanistic study
Authors: Noor U Din Reshi, Lakshay Kathuria, and Ashoka Samuelson
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Inorg. Chem. 10.1002/ejic.201900455
Link to VoR: http://dx.doi.org/10.1002/ejic.201900455

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Chemoselective reduction of imines catalyzed by Ru(II)
half-sandwich complexes: a mechanistic study
(a)
(a)
(a)
Noor U Din Reshi , Lakshay Kathuria , Ashoka G. Samuelson *
Abstract
Ruthenium half-sandwich complexes ligated to in the reduction of N-benzylidineaniline using
silyl hydrides as reductants has been examined. The chemoselective reduction of imines takes
place under mild conditions to afford the corresponding amines in nearly quantitative yield.
Mechanistic studies indicate that dissociation of the ancillary ligands generate the active
catalyst in all the complexes studied, which is the same species generated by [Ru(p-
cymene)(Cl)
2
2
] under the reaction conditions. This results in the formation of a single
catalytic species irrespective of the starting half-sandwich complex. Detailed mechanistic
studies involving trapping of intermediates, in situ studies using mass spectrometry and NMR
spectroscopy were carried out using the active catalyst generated by [Ru(p-cymene)(Cl)
2
2
] .
The mechanism of the reaction is dependent on the number of the hydrogen atoms in the
reducing silane. The reaction proceeds via Gade-Hoffman pathway or Zheng-Chan pathway
when a dihydro or trihydrosilane is the reductant. However, the use of a monohydrosilane,
leads to longer reaction times presumably due to a change in the reaction pathway.
Keywords:
Reduction
Hydrosilylation
Silyl hydrides
Ru(II) half-sandwich complexes
Imines
(a)
Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore
60012, Karnataka, India.
5
*
Corresponding author: Ashoka G. Samuelson.
*
E-mail: ashoka@iisc.ac.in; Academic title: Prof.; Homepage: http://ipc.iisc.ac.in/~ags/
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Introduction
Amines are widely used in the synthesis of agrochemicals, natural products, pharmaceuticals,
as well as biologically valuable fine chemicals.[ They are present in different alkaloids,
amino acids, and nucleotides. Several methodologies have been developed to synthesize
amines. For example, the transition-metal catalyzed C-N bond formation especially using Pd-
catalyzed Buchwald–Hartwig and Cu-catalyzed Ullman reactions has emerged as a useful
1–4]
[5,6]
method in amine synthesis.
The use of stoichiometric quantities of reductants such as
lithium aluminium hydride and lithium borohydride for reducing imines, amides, cyanides,
and nitroarenes is an alternative route to synthesize amines. However, these reductants are
sensitive to air and moisture, incompatible to various functional groups and result in the
formation of metal salt by-products.[ Although the alkylation of ammonia using alkyl
halides or alcohols is a simple method for preparation of amines, alkylation is difficult to
control with alkyl iodides and bromides.[
7]
8]
The transition metal catalyzed hydrogenation, transfer hydrogenation, and reductive
amination are also simple methods for amine synthesis. Although hydrogen is the ideal
reducing agent, it is not suitable for many substrates due to high temperatures and high
[
9–12]
pressures required, and absence of chemoselectivity.
Reductive amination of carbonyl
compounds using primary amines, via sequential condensation and catalytic reduction,
constitutes a basic method for producing substituted amines, but it often requires drastic
[
13–19]
reaction conditions.
Transfer hydrogenation using silanes provides an attractive
alternative method of reduction through hydrosilylation.
Hydrosilylation of imines has been performed under mild conditions in the presence of
various metal catalysts; however, in comparison with the asymmetric hydrosilylation of
prochiral ketones, the reaction of prochiral imines has been more challenging in achieving
[
20]
good enantioselectivity.
Most of the homogeneous transition metal catalysts for
hydrosilylation of imines are based on platinum, rhodium, palladium, and titanium[
.
21]
However, catalysts based on ruthenium, iridium, iron, zinc, nickel, palladium, and rhenium
[
22]
are also reported.
Catalysts based on Ru are widely used in reduction, more so in
asymmetric hydrogenation and transfer hydrogenation of imines.[
there are not many examples of ruthenium catalyzed hydrosilylation of imines.[
23–25]
On the other hand,
26–30]
Dixneuf and co-workers reported [RuCl (p-cymene)] (P) as an efficient catalyst for the
2
2
[31]
hydrosilylation of imines. They obtained a variety of secondary amines in 69–97% yield
from the corresponding imines using a catalyst loading of 1-2 mol%. The reaction was done
in ethanol under ambient conditions at room temperature using PMHS as reductant which is
an inexpensive by-product from the silicon industry. Furthermore, this strategy offered a very
good chemoselectivity for reduction of the C=N bond and functional groups like -CN, -NO ,
2
C=O, and C=C were tolerated. Furthermore, the use of ethanol eliminated the need for a
desilylation step which is usually involved in the hydrosilylation reactions to generate the
product from the intermediate silylamine or silylether. This method was also applied
successfully for hydrosilylation of difficult to reduce ketimines using 2 mol% of the catalyst
at room temperature for 4–20 h.
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
In this work, we examined the catalytic activity of ruthenium half sandwich complexes 1a-1g
and 2a-2e (Figure 1) based on chiral 2-oxazolidinethiones and 2-thiozolidinethiones as
ligands for their use in the hydrosilylation of imines. Initial mechanistic studies indicated that
the catalytically active intermediate generated by these thione coordinated catalysts was
similar to that generated by the achiral, Dixneuf catalyst, [Ru(p-cymene)(Cl)
2
] . This resulted
2
in the formation of a racemic amine when prochiral ketimines were reduced. The detailed
reaction path adopted by [Ru(p-cymene)(Cl)
2
2
] to catalyse selective reduction of C=N bond
was also studied.
Results and discussion
Synthesis
Our efforts to generate an enantioselective catalyst for the synthesis of amines led us to chiral
half-sandwich ruthenium complexes shown below (see Fig. S1 for complete list). They could
be readily made from chiral 2-oxazolidinethiones and 2-thiozolidinethiones which in turn
were prepared using a literature procedure from natural amino acids and the cymene dimer
[32]
[
Ru(p-cymene)(Cl)
2
2
] .
X = S;
R
isopropyl 1a
isobuyl
Ru
Cl
Cl
S
1b
HN
X
sec. butyl 1c
R
Figure 1. Ruthenium half sandwich complexes studied in this work (see also Fig. S1).
Reduction of imines
In order to test the catalytic activity of the complexes in the reduction of aldimines a test
substrate N-benzylidineaniline was used. The optimization studies carried out using the
-1
representative complex 1a showed that the best turn over frequency (TOF = 400 h ) was
obtained using ethanol as the solvent. Moreover, the use of ethanol eliminated the need for
[
31]
the desilylation step to get amines. Subsequently, all the complexes were tested using the
optimized conditions. The complexes showed excellent catalytic activity under the optimized
conditions and the catalytic activity was independent of the ancillary ligand resulting in the
complete reduction of N-benzylidineaniline in 15 minutes (Scheme 1, R = H; Supporting
Information Table S1).
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Scheme 1. Catalytic reduction of imines
R
R
Catalyst (1-2 mol%), silane (2 equiv.)
Ph
Ph
Ph
N
Ph
N
ethanol, room temperature
H
R = H, CH3
Further optimization of the reaction conditions involved different reagents and the optimum
amounts of the reducing agent needed for the reaction. Interestingly, when 0.34 equiv. of
imine was used, the reaction became significantly slower and the conversion decreased
marginally to 90 % (Entry 3, Table 1). This decrease in the rate of the reaction could be due
to a slower hydride transfer from intermediate sterically crowded hydride donors
(
C
2
H
5
O)
rate of reaction when a dihydrosilane (diphenylsilane) was used instead of phenylsilane
Entry 2, Table 1). However, the rate of the reaction decreased significantly when a
n
SiH3-nPh as compared to PhSiH . There was little effect on the conversion and the
3
(
monohydrosilane, triphenylsilane, was used (Entry 5, Table 1). The slower reactivity of
monohyrdosilanes as compared to trihydrosilanes and dihydrosilanes could also be due to the
operation of different mechanisms vide infra.
Table 1. Effect of varying the nature and amount of silane.
Ru
Cl
S
Cl
HN
X
1
c (1 mol%), silane
Ph
Ph
Ph
N
Ph
N
ethanol, room temperature
H
0.25 mmol
Conversion
Entry
Silane
PhSiH
Ph SiH
PhSiH
Ph SiH
Silane (equiv.)
Time (min)
a
(%)
1
2
3
4
5
6
3
2
2
15
15
30
30
30
60
100
100
91
2
2
3
0.34
0.5
1
2
2
92
Ph
Ph
3
SiH
SiH
38
3
1
91
a
1
By H NMR.
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
A primary reason for testing these chiral thione complexes as catalysts was to probe the
asymmetric reduction of ketimines under the conditions used for reducing aldimines so
effectively. Thus, all complexes afforded nearly quantitative yields in 2 h for the reduction of
N,1-diphenylethan-1-imine using 2 mol% of any of these complexes (Scheme 1, R = CH ;
3
See Supporting Information Table S2). Surprisingly, the ee was zero with all the complexes.
o
When the reaction was carried out at 0 C, the reaction took 6 h to complete but the ee still
remained zero (Table S2). Since the complexes showed very similar catalytic activity, only
complex 1c was tested as a catalyst for the reduction of various imines (See Supporting
Information Table S4). However, no chirality induction was observed.
Although ruthenium(II) catalysts have been used for the hydrosilylation of alkenes[33], and
ketones[26,34–36], this ruthenium(II) hydrosilylation of imines is very chemoselective similar to
that observed by Dixneuf group which used the ruthenium dimer P. Thus, using this catalyst,
acetophenone, and styrene could not be reduced. All observations regarding the effect of the
nature of the substituent on the reactivity and chemoselectivity towards hydrosilylation are
similar to that observed using [(η6-cymene)RuCl
2
2
] (P) as the catalyst strongly pointing to a
common intermediate either P or one generated by P and the complexes studied by us.
Mechanistic studies
The extremely similar reactivity of Ru half-sandwich complexes of thiones and the complete
absence of enantioselectivity led to a suspicion that these complexes generate a common
intermediate under the reaction conditions. Confirmation of the common intermediate under
the reaction conditions needed some more studies with the Ru half-sandwich complexes of
1
thiones using NMR and mass spectrometry. The H NMR spectrum of 1c in CDCl
3
indicated
1
the presence of a single compound. However, its H NMR spectrum in CD
3
OD showed the
6
presence of [(η -cymene)RuCl
2
] (P) , free ligand (L) and a new compound (T) and the ratio
2
of P: L: T was estimated to be 1: 2: 4. The excellent catalytic activity of P, which does not
contain the ancillary ligand, in the hydrosilylation of imines to the corresponding amines in
[
31]
ethanol is known.
As observed earlier, the catalytic reactivity pattern of Ru-thione
complexes studied in this work matches that of P. Furthermore, the formation of P from 1c in
alcohols (deuterated methanol) points to its involvement in the reaction carried out in ethanol.
To test the hypothesis that P was the actual catalyst generated in situ, a test reaction was
carried out using N-benzylidineaniline as the substrate and 1c (1 mol%) as the catalyst in
ethanol in the presence of added thione ligand (5 mol%) in ethanol before the addition of
substrate and phenylsilane (2 equiv.). This led to a dramatic decrease in the catalytic activity
and only 20 % of N-benzylidineaniline was reduced in 4 h. The suppression of the
dissociation of the thione from ruthenium in complex 1c by adding the ligand L1c decreases
the amount of P and hence reduces the catalytic activity making the rate comparable to the
corresponding complex.
The intermediate T is an unsymmetrical complex as indicated by the presence of four sets of
peaks in the cymene region. The ratio of arene to ancillary ligand (L) in T is 1:1 as indicated
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
by the integration ratios of the corresponding peaks. It was impossible to elucidate the
1
structure of the complex formed in methanol purely from the H NMR spectrum although
mass spectrometry was more helpful vide infra. However, Ru half-sandwich complexes of
[
37–39]
heterocyclic thiones are known to form dimers under similar conditions.
For example
6
[
Ru(η -cymene)Cl
2
(2-mercaptobenzothiazole)] forms a dinuclear species in the mixture of
[
37]
methanol and water (10% methanol).
Taking a cue from these reports, the dinuclear
species is assigned a plausible structure T (Scheme 2). However, it is likely that T is a mere
bystander as there was no chirality induction observed in the reduction of prochiral imines.
The positive-ion mode ESI-MS spectra of 1c showed peaks at 410.0553 and 855.0792 that
-
+
-
+
were assigned respectively to [1c-(HCl
2
) ]
and the dinuclear species [T-Cl ] . The
formation of P and the dissociation of the ancillary ligand from the Ru centre was indicated
by the presence of peaks at 576.9561 and 294.0440 which were assigned respectively to P-
-
6
+
Cl and [(η -cym)Ru(OH)(CH
3
CN)] . There was no indication of the loss of the arene ring in
1
the H NMR spectrum and ESI-MS spectrum. These observations were similar to those made
with Ru-half-sandwich complexes of monodentate S coordinating ligands such as 2-
mercaptobenzothiazole,
[
37]
2-mercapto-1-methylimidazole,
and
4,5-diphenyl-2-
[
39]
mercaptoimidazole
.
Scheme 2. Formation of dinuclear complex T from 1c in MeOH
Cl
S
N
Cl
MeOH
2 HCl
S
Ru
Ru
Ru
2
-
Cl
S
S
Cl
S
HN
N
S
T
1c
S
S
HN
L1c =
6
The achiral metal complex [(η -cymene)RuCl
2
] (P) formed in situ was a better catalyst, all
2
further mechanistic studies were carried out with P.
P (0.008 mmoles), trimethoxybenzene (0.75 mg, 0.0044 mmoles as internal standard) and
CDCl
3
(0.4 mL) were loaded in an NMR tube. Diphenylsilane (0.026 mmoles) was added and
1
the reaction was followed by H NMR spectroscopy. A small amount of a gas was released
immediately after the addition of diphenylsilane. The evolution of gas was also observed
during the catalytic reactions on the addition of the silane to the solution containing the
catalyst and substrate. Examination of the NMR spectrum recorded immediately after the
addition of diphenylsialne showed a singlet at δ = 4.62 ppm, which can be assigned to
dihydrogen. The release of dihydrogen can be rationalized by the reactions shown in Scheme
3
. The reductive elimination of the dihydride species (Intermediate X) to furnish Y was
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
examined by DFT calculations using a slightly truncated model and found to be
thermodynamically accessible (ΔG = 9.47 kcal/mol; L: HN=CH ; see Supporting
2
Information).
Scheme 3. Formation of A
A Ru-dihydride complex formed by the sigma bond metathesis can lose dihydrogen by the
reductive elimination. The Ru(0) intermediate thus generated can oxidatively add a molecule
1
of diphenylsilane to form species A. The presence of A was inferred from the H NMR
spectrum recorded immediately after the addition of diphenylsilane. The peaks at δ = -10.17
ppm and 5.74 ppm were assigned respectively to the Ru-H and Ru-SiH protons in A. Similar
[
33,40–45]
1
chemical shifts have been assigned to Ru-H and Ru-SiH protons.
The H NMR
spectrum also showed the presence of an unidentified Ru-cymene complex (B) besides A,
1
and P. After 30 minutes, P and diphenylsilane completely disappeared from the H NMR
spectrum. The species A and B were present in the ratio A: B = 4.25:1 after 30 minutes and
the total amount of A and B was estimated to be close to 95% of P used. After 30 minutes
when dominant species present in the reaction mixture was A, the substrate imine S was
added. (1 equiv. w.r.t A, 2.6 mg). After the addition of S, peak at δ = 5.74 ppm (SiH) started
decreasing in intensity along with the peak at δ = 8.53 ppm (CH peak of imine).
Concomitantly a peak appeared at δ = 4.37 ppm. The latter resonance might be assigned to
the CH of the reduced amine. In 4 h, the peaks at δ = 5.74 and δ = 8.53 completely
disappeared. The new species C was assigned the structure containing “Ru(H)(silylamine)”
(Scheme 4). Interestingly, species C has a hydride peak at δ = -10.17 ppm. It indicated that
the imine did not insert into the metal-hydride bond which is the path followed in the
hydrosilylation of olefins catalyzed by the late transition metal complexes (Chalk-Harrod
[
46]
mechanism).
7
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Scheme 4. Reaction of imine with A
N
R
Ph SiH2
2
Ru
Ru
Ru
N
H
H
H
L
L
Ph HSi
2
L
Ph HSi
2
Ph Si
2
A
A
Ph HSi
2
H
R
N
C
O
H
R
N
S =
=
N
R
When 5 µL of diphenylsilane was added to the NMR tube containing C, A was regenerated
Scheme 4). This can happen via the reductive elimination of silylamine from C, followed by
(
the oxidative addition of diphenylsilane.
Scheme 5 shows one plausible mechanism for this reaction whereby the formation of A is
followed by the insertion of the substrate into M−Si bond. The subsequent reductive
[
47]
elimination yields the respective silyl amine (Ojima mechanism). Similar to many reports,
the use of monohydrosilanes led to slower reaction rates relative to the use of dihydrosilanes
[
48–53]
or trihydrosilanes (Table 1).
The rate enhancement can be due to the effect of steric
properties of silane. Tertiary silanes being sterically bulky is likely to transfer a hydride
slower than a secondary silane which in turn should transfer hydride slower than a primary
silane. However, the similar rates of primary and secondary silanes (Ph
3
SiH < Ph
2
SiH =
2
PhSiH ) cannot be explained purely by steric factors. The Zheng-Chan mechanism (ZC
3
mechanism) and the Gade-Hoffman silylene mechanism (GH mechanism) (Scheme 6) are
not accessible to a tertiary silane.
8
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Scheme 5. Ojima-like reaction mechanism
N
L
R
R
HC
Ru
Ru
Ru
HC R
H
H
H
N
L
N
Ph HSi
2
Ph2HSi
L
L
SiHPh2
A
2
SiH Ph2
reductive elimination
C H OH
2
5
R
HN
Ru
H C
2
H
N
R
L
SiHPh2
(
C H O)SiHPh
2 5 2
In the ZC pathway, the imine binds to the electrophilic Si center and the hydrogen atom on Si
is transferred to the substrate via a four-membered transition state to form C. Whereas in GH
pathway, the hydrogen atom on Si is transferred to ruthenium to form a dihydride silylene
intermediate. The generation of silylene complexes by the double Si-H activation pathway
has been proposed for many rhodium and ruthenium complexes during the hydrosilylation
[
50,54]
reactions.
After the substrate binds to the Si center, one of the hydride ligands is
transferred to the substrate via a five-membered transition state to form C.
Scheme 6. Possible Zheng-Chan mechanism (top) and Gade-Hoffman mechanism
(bottom) for the reaction
Ru
ZC pathway
+ imine
+ L
Ru
Ru
N
Ru
Ru
N
Ru
N
-
L
H
H
H
H
H
Ph Si
2
H
Ph Si
2
L
Ph HSi
2
L
Ph HSi
2
Ph HSi
2
SiPh2
N
H
A
H
H
R
R
R
GH pathway
R
C
+
imine
Ru
Ru
Ru
H
H
H
SiPh2
H
SiPh2
N
H
SiPh2
H
N
R
R
Since there was no direct experimental evidence for the formation of an intermediate silylene,
[
52,54]
a trapping experiment similar to that reported by Kuhn et al.
To 5 mg of P dissolved in 0.5 mL of CDCl , 25 μL (20.7 mg, 0.15 mmol) of t-BuMe
and 50 μL (50.1 mg, 0.27 mmole) of Ph SiH
was performed to detect it.
3
2
SiOH
1
2
2
were added. The reaction was monitored by H
NMR and Si NMR spectroscopy. After 60 minutes, silanol completely disappeared from the
2
9
1
H NMR spectrum with the formation of t-BuSiMe
2
-O-SiHPh
2
. The disiloxane was
9
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
1
29
[52]
characterized by H NMR and Si NMR spectroscopy. Disiloxane can be formed by the
coordination of the silanol to the electrophilic Si centre of the Ru-silylene complex, which is
followed by the loss of H
2
and reductive elimination (Scheme 7, Path A). Alternatively,
disiloxane can be formed by the reaction of monohydride complex A with the silanol
(
Scheme 7, path B). In the latter case, a monohydrosilane is expected to show a similar
[
52]
reactivity as that of a dihydrosilane.
and Ph
However, there was no reaction between the silanol
SiH: P) up to 12 hr, hence the formation of
3
SiH in presence of P (silanol: Ph
3
disiloxane by the reaction of A with silanol (Path B) can be ruled out.
Scheme 7. Pathways for the formation of disiloxane
Path B
silanol
+
L
+
Ru
Ru
-
HH intermolecular,
concerted
Ph HSi
Ph HSi
2
2
H
H
O
A
H
Si
H
SiPh2
O
+
Si
Ru(0)
L
reductive elimination
Path A
silanol
+
Ru
Ru
Ru
H
H
-
HH
H
+
L
H
H
SiPh2
SiPh2
SiPh2
H
O
O
Si
Si
The formation of alkoxysilanes Ph4-nSi(OC
step was also confirmed from the H NMR spectra of crude catalytic reaction mixtures.
2
H
5
) which are reductively eliminated in the final
n
1
To further differentiate between ZC and GH mechanisms, the reaction was investigated using
computational tools. The reaction intermediates, using a slightly truncated model of [Ru(p-
cymene)H(SiHPh
2
)] which is probably the actual catalyst were examined using DFT
was used as the model silane and methanimine as the model imine. Key
calculations. Ph SiH
2
2
catalytic steps were calculated at the B3LYP/LanL2DZ(Ru)/6-311G (H,C,N,O,Si) level of
theory. (Scheme 8). The imaginary frequency, which characterizes the transition state Z2,
represents a vibrational motion, which leads to the breaking of Si-H bond and the formation
of C-H bond. The corresponding transition state in the GH mechanism (S3) which is
characterized by the imaginary frequency that leads to the transfer of hydride ligand to the
unsaturated carbon was also located. Interestingly, S3 is lower in energy relative to Z2 by 28
kcal/mol. The hydride migration from silyl group to Ru in A′ leads to the formation of
silylene intermediate (S1) with a Ru=Si bond (2.24 Å) which is significantly shorter than Ru-
1
0
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
Si bond in A′ (2.38 Å). Due to the very steep nature of the potential energy surface, all the
efforts to locate the transition state connecting S1 to S2 failed. However, the large difference
in the energies of Z2 and S3 is probably a decisive factor to favor the GH pathway.
Scheme 8. Comparison of the computed free energy profiles (B3LYP/LanL2DZ(Ru)/6-
3
11G (H,C,N,O,Si) of the ZC (blue) and GH pathway (red) using methanimine as
substrate
#
ZC pathway
imine
Ru
Ru
Ru
N
+
Ru
H
H
H
H
Ph HSi
2
SiPh2
N
Ph Si
2
H
H
H
Ph HSi
2
N
A'
H
H
Z1
Z2
C'
GH pathway
#
+
imine
Ru
Ru
Ru
H
H
H
SiPh2
H
SiPh2
H
SiPh2
H
S1
H
N
N
H
S2
S3
Based on these observations this reaction proceeds through a silylene intermediate similar to
the Gade-Hoffman mechanism (Scheme 9).[
50,52]
In the case of monohydrosilanes, such as
1
1
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
triphenylsilane where ZC and GH mechanisms are not accessible, the reaction can take place
via a slower pathway (Ojima mechanism).
Scheme 9. Proposed mechanism
+
-
Ph SiH2
2
2
Cl
Ru
Cl
Ru
Ru(0)
2
Ph SiCl2
H
H2
L
L
L
H
Ph SiH
2
2
Ph HSi
2
C H OH
HN
2
5
Ru
N
-
L
- (C H O)SiHPh
H
H
2
5
2
R
H
L
Ph SiH
R
Ph HSi
2
2
2
A
Ru
H
Ru
N
H
SiPh2
H
Ph Si
L
2
H
R
+
imine
C
Ru
H
SiPh2
H
+ L
N
R
Ru
H
H
SiPh2
N
R
Conclusions
All half-sandwich complexes proved equally good in the chemoselective reduction of
1
aldimines and ketimines. H NMR spectroscopy suggested that the common catalyst
generated by Ru-thione complexes is the same species generated by [Ru(p-cymene)(Cl)
2
2
] .
This helps one to understand the observations such as the drastic decrease in the catalytic
activity by adding excess ligand, no enantiomeric excess in the reduction of prochiral
substrates, and similar reactivity pattern of different Ru half-sandwich complexes to that of P.
The rate enhancement observed with di and trihydrosilanes along with the results of various
stoichiometric reactions can be explained by both ZC and GH pathways. Furthermore, the
trapping experiment gave an indirect evidence for the formation of a silylene intermediate.
However, ZC pathway could not be completely ruled out based on DFT calculations. In case
1
2
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
of a monohydrosilanes such as triphenylsilane where ZC and GH mechanisms are not
accessible, the reaction can take place via a slower pathway presumably like the mechanism
proposed by Ojima.
Experimental section
General remarks
All catalytic reactions and manipulations were carried under ambient conditions. HPLC grade
ethanol was used. Analytical thin layer chromatography (TLC) was performed using
precoated silica gel plates supplied by Merck (60 F254). After elution, the plates were
visualized using UV radiation (254 nm). Preparative TLC was done using ChemLabs Silica
gel GF 254. The NMR spectra were recorded on a Bruker AMX 400 spectrometer that
1
13
1
operates at 400 MHz for H and 100 MHz for C NMR. Chemical shifts for H NMR spectra
are reported in ppm relative to the chloroform signal at δ 7.26 as a singlet.
Synthesis
Ru half-sandwich complexes of chiral 2-oxazolidinethiones and 2-thiozolidinethiones were
prepared according to a literature procedure.[ Aldimines were prepared similar to the
reported procedure, by refluxing an equivalent mixture of amine and aldehyde in dry ethanol
32]
[
55]
in the presence of a catalytic amount of acetic acid and molecular sieves. Ketimines were
prepared by refluxing an equivalent mixture of amine and ketone in dry xylene in presence of
catalytic amount of p-tolunesulphonic acid and molecular sieves.[ Both aldimines and
ketimines were purified by recrystallization.
56]
General procedure for reduction
The catalyst (0.0025 mmol) was dissolved in 5 mL of spectroscopic grade ethanol. 0.25
mmol of imine was added to this solution. This was followed by the addition of PhSiH (0.5
3
mmol). Then the reaction mixture was stirred at room temperature. Aliquots from the reaction
mixture were withdrawn at specified intervals. The volatiles were removed using a rotary
1
evaporator. Then HNMR spectra of the residues were recorded in CDCl
3
. Conversion was
calculated from the intensity ratios of the peaks arising from the reactant and product. The
amines were purified by preparative thin layer chromatography using a mixture of
hexane/ethyl acetate (2-5%) as the eluent. The enantiomeric excess was determined using
optical activity measurements.
Characterization of A
1H NMR (CDCl3, 400 MHz): δ = -10.17 (1H, s, hydride), 1.41 (3H, d, J = 6.8 Hz, isopropyl),
1
.44 (3H, d, J = 6.8 Hz, isopropyl), 2.29 (3H, s, methyl), 2.99 (1H, m, isopropyl), 4.85 (1H, d,
J = 5.9 Hz, cymene), 5.30(1H, d, J = 5.9 Hz cymene), 5.54(1H, d, J=5.9 Hz, cymene),5.64
1H,d,J=5.9 Hz, cymene), 5.74 (1H, s, SiH), 7-8 ( 10H, SiHPh ).
(
2
Characterization of C
1
3
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
1H NMR (CDCl3, 400 MHz): δ = -10.17 (1H, s, hydride), 1.40 (3H, d, J = 7 Hz, isopropyl),
.44 (3H, d, J = 7 Hz, isopropyl), 2.29 (3H, s, methyl), 2.98 (1H, m, isopropyl), 3.72 (3H, s,
OCH ), 4.84 (1H, d, J = 5.9 Hz, cymene), 5.30(1H, d, J = 5.9 Hz cymene), 5.54(1H, d, J=5.9
1
3
Hz, cymene),5.64 (1H,d,J=5.9 Hz, cymene), 7-8 ( 10H, SiHPh
amine).
2
), 6-8 (10H Ph of bound
2
-O-SiHPh [
52]
Characterization of t-BuSiMe
2
2
1H NMR (CD
Cl
, 400 MHz) δ (ppm) = 0.08 (s, 6H, SiCH
), 0.90 (s, 9H, CCH
Cl , 79 MHz) δ = -22.5 (s,
), 5.53 (s, 1H,
2
2
3
3
2
9
SiH), 7.36 (m, 6H, Ar-H),7.59 (m, 4H, Ar-H); Si NMR (CD
2
2
Ph HSiO), 15.0 (s, t-BuMe SiO).
2
2
Computational details
Geometry optimizations were carried out using the suite of programs available in the
[
57]
[58]
Gaussian09 (G09) package
using the DFT B3LYP method.
All optimizations and
vibrational frequency calculations were performed in the gas phase without symmetry and
[
59,60]
geometry constraints. The 6–311G
basis set was used for main group elements and
[61–63]
hydrogen. The LANL2DZ basis set with effective core potential (ECP) was used for Ru.
Transition states were located using the Berny algorithm.[
64,65]
The number of imaginary
frequencies (Nimag) obtained for the optimized structures was used to characterize the
geometry minima (Nimag = 0) and transition states (Nimag = 1). Intrinsic reaction coordinate
calculations were carried out to confirm that the transition state structures connect the
[
66]
reactants and the respective products.
Acknowledgments
N.U.R thanks Council of Scientific and Industrial Research (CSIR), New Delhi for senior
research fellowship. Authors acknowledge Department of Science and Technology (DST),
New Delhi for research grant (DSTO-1507/2017-1937).
Conflict of Interest
There are no conflicts to declare
References:
[
[
[
1] A. Ricci, Amino Group Chemistry. From Synthesis to the Life Sciences, Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, 2008.
2] S. A. Lawrence, Amines: Synthesis, Properties and Applications, Cambridge
University Press, Cambridge, 2004.
3] A. Ricci, Modern Amination Method, Wiley-VCH Verlag GmbH & Co. KGaA,
Weinheim, New York, 2000.
1
4
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
[
4] R. C. Larock, Comprehensive Organic Transformations: A Guide to Functional Group
Preparations. Two Volume Set, 2nd Edition, 2011.
[
[
[
5] J. F. Hartwig, Acc. Chem. Res. 2008, 41, 1534–1544.
6] D. S. Surry, S. L. Buchwald, Angew. Chemie - Int. Ed. 2008, 47, 6338–6361.
7] J. Seyden-Penne, Reductions by the Alumino- and Borohydrides in Organic Synthesis,
WILEY-VCH New York, 1997.
[
[
[
8] K. Eller, E. Henkes, R. Rossbacher, Ullmann’s Encycl. 2002.
9] Stein, Mario, B. Breit., Angew. Chemie Int. Ed. 2013, 52, 2231-2234.
10] G. Beamson, A. J. Papworth, C. Philipps, A. M. Smith, R. Whyman, J. Catal. 2010,
2
69, 93–102.
[
[
[
11] G. Beamson, A. J. Papworth, C. Philipps, A. M. Smith, R. Whyman, Adv. Synth. Catal.
2
010, 352, 869–883.
12] A. A. Núñez Magro, G. R. Eastham, D. J. Cole-Hamilton, Chem. Commun. 2007, 58,
3
154–3156.
13] A. Pagnoux-Ozherelyeva, N. Pannetier, M. D. Mbaye, S. Gaillard, J. L. Renaud,
Angew. Chemie - Int. Ed. 2012, 51, 4976–4980.
[
[
14] O. Bondarev, C. Bruneau, Tetrahedron Asymmetry 2010, 21, 1350–1354.
15] D. Steinhuebel, Y. Sun, K. Matsumura, N. Sayo, T. Saito, J. Am. Chem. Soc. 2009,
1
31, 11316–11317.
[
[
16] V. I. Tararov, A. Börner, Synlett 2005, 203–211.
17] V. I. Tararov, R. Kadyrov, T. H. Riermeier, C. Fischer, A. Börner, Adv. Synth. Catal.
2
004, 346, 561–565.
[
[
18] S. Gomez, J. A. Peters, T. Maschmeyer, Adv. Synth. Catal. 2002, 344, 1037–1057.
19] S. Nishimura, Handbook of Heterogeneous Catalytic Hydrogenation for Organic
Synthesis, John Wiley & Sons, Inc., New York, 2001.
[
[
[
[
20] I. Ojima, The Chemistry of Organic Silicon Compounds, Wiley, Chichester, 1989.
21] J. Yun, S. L. Buchwald, J. Am. Chem. Soc. 1999, 121, 5640–5644.
22] B. Li, J.-B. Sortais, C. Darcel, RSC Adv. 2016, 6, 57603–57625.
23] T. Ohkuma, H. Ooka, S. Hashiguchi, T. Ikariya, R. Noyori, J. Am. Chem. Soc. 1995,
1
17, 2675–2676.
[
24] T. Ohkuma, H. Doucet, T. Pham, K. Mikami, T. Korenaga, M. Terada, R. Noyori, J.
Am. Chem. Soc. 1998, 120, 1086–1087.
[
[
25] R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97–102.
26] Y. Nishibayashi, I. Takei, S. Uemura, M. Hidai, S. Uemura, Organometallics 1998, 17,
1
5
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
3
420–3422.
[
27] I. Takei, Y. Nishibayashi, Y. Ishii, Y. Mizobe, S. Uemura, M. Hidai, Chem. Commun.
2
001, 83, 2360–2361.
[
[
[
[
[
28] H. Hashimoto, I. Aratani, C. Kabuto, M. Kira, Organometallics 2003, 22, 2199–2201.
29] S. Abbina, S. Bian, C. Oian, G. Du, ACS Catal. 2013, 3, 678–684.
30] J. Hermeke, H. F. T. Klare, M. Oestreich, Chem. - A Eur. J. 2014, 20, 9250–9254.
31] B. Li, J. B. Sortais, C. Darcel, P. H. Dixneuf, ChemSusChem 2012, 5, 396–399.
32] N. U. D. Reshi, S. Dineshchakravarthy, A. G. Samuelson, Appl. Organomet. Chem.
2
019, 33, e4885.
[
[
33] T. Tuttle, D. Wang, W. Thiel, J. Kohler, M. Hofmann, J. Weis, Dalt. Trans 2009, 0,
5
894–5901.
34] C. Song, C. Ma, Y. Ma, W. Feng, S. Ma, Q. Chai, M. B. Andrus, Tetrahedron Lett.
2
005, 46, 3241–3244.
[
[
[
[
35] C. Moreau, C. G. Frost, B. Murrer, Tetrahedron Lett. 1999, 40, 5617–5620.
36] G. Zhu, M. Terry, X. Zhang, J. Organomet. Chem. 1997, 547, 97–101.
37] R. Mitra, A. G. Samuelson, Eur. J. Inorg. Chem. 2014, 2014, 3536–3546.
38] R. Mitra, S. Das, S. V. Shinde, S. Sinha, K. Somasundaram, A. G. Samuelson, Chem. -
A Eur. J. 2012, 18, 12278–12291.
[
[
[
[
39] S. Sheldrick, Inorganica Chim. Acta 1993, 208, 145–151.
40] B. Li, J.-B. Sortais, C. Darcel, Chem. Commun. 2013, 49, 3691.
41] K. Umezawa-Vizzini, T. R. Lee, Organometallics 2004, 23, 1448–1452.
42] V. K. Dioumaev, B. R. Yoo, L. J. Procopio, P. J. Carroll, D. H. Berry, J. Am. Chem.
Soc. 2003, 125, 8936–8948.
[
[
[
43] V. Montiel-Palma, R. N. Perutz, M. W. George, O. S. Jina, S. Sabo-Etienne, Chem.
Commun. 2000, 0, 1175–1176.
44] G. Meister, G. Rheinwald, H. Stoeckli-Evans, G. Süss-Fink, J. Organomet. Chem.
1
995, 496, 197–205.
45] J. Depasquale, M. Kumar, M. Zeller, E. T. Papish, Organometallics 2013, 32, 966–
9
79.
[
[
46] A. J. Chalk, J. F. Harrod, J. Am. Chem. Soc. 1965, 87, 16--21.
47] I. Ojima, M. Nihonyanagi, T. Kogure, M. Kumagai, S. Horiuchi, K. Nakatsugawa, Y.
Nagai, J. Organomet. Chem. 1975, 94, 449–461.
[
48] I. Ojima, T. Kogure, M. Kumagai, S. Horiuchi, T. Sato, J. Organomet. Chem. 1976,
1
22, 83–97.
1
6
This article is protected by copyright. All rights reserved.

European Journal of Inorganic Chemistry
10.1002/ejic.201900455
[
[
49] G. Z. Zheng, T. H. Chan, Organometallics 1995, 14, 70–79.
50] N. Schneider, M. Finger, C. Haferkemper, S. Bellemin-Laponnaz, P. Hofmann, L. H.
Gade, Angew. Chemie - Int. Ed. 2009, 48, 1609–1613.
[
[
51] I. Ojima, M. Nihonyanagi, T. Kogure, M. Kumagai, S. Horiuchi, K. Nakatsugawa, Y.
Nagai, J. Organomet. Chem. 1975, 94, 449–461.
52] P. Gigler, B. Bechlars, W. A. Herrmann, F. E. Kühn, J. Am. Chem. Soc. 2011, 133,
1
589–1596.
[
[
53] N. Schneider, M. Finger, C. Haferkemper, Chem. Eur. J. 2009, 15, 11515–11529.
54] M. C. Lipke, A. L. Liberman-Martin, T. D. Tilley, Angew. Chemie - Int. Ed. 2017, 56,
2
260–2294.
[
[
[
55] R. Montalvo-Gonzalez, A. Ariza-Castolo, J. Mol. Struct. 2003, 655, 375–389.
56] T. Chen, X. Chen, J. Wei, D. Lin, Y. Xie, W. Zeng, Org. Lett. 2016, 18, 2078–2081.
57] J. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, H. R.
Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, G. Nakatsuji, M.
Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, J. Zheng, J. L. Sonnenberg,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, T. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, J. J. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, R. K. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T.
Keith, J. J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. B. C.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, O. V. Bakken, C.
Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, et al., 2013.
[
[
[
[
[
[
[
[
[
58] A. P. Scott, L. Radom, J. Phys. Chem. 1996, 100, 16502–16513.
59] W. J. Hehre, K. Ditchfield, J. A. Pople, J. Chem. Phys. 1972, 56, 2257–2261.
60] J. Stephen Binkley, J. A. Pople, W. J. Hehre, J. Am. Chem. Soc. 1979, 102, 939–947.
61] A. D. Becke, J. Chem. Phys. 1985, 82, 285–298.
62] W. J. Hehre, K. Ditchfield, J. A. Pople, J. Chem. Phys. 1985, 82, 299–310.
63] R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1985, 82, 270–283.
64] X. Li, M. J. Frisch, J. Chem. Theory Comput. 2006, 2, 835–839.
65] C. Peng, H. Bernhard Schlegel, Isr. J. Chem. 1993, 33, 449–454.
66] H. P. Hratchian, H. B. Schlegel, J. Chem. Theory Comput. 2005, 1, 61–69.
1
7
This article is protected by copyright. All rights reserved.