Efficient and Selective Hydrosilylation of Secondary and Tertiary Amides Catalyzed by an Iridium(III) Metallacycle: Development and Mechanistic Investigation

Article abstract of DOI:10.1002/cctc.201700400

Readily accessible cationic Ir^III metallacycles catalyze efficiently the chemoselective hydrosilylation of tertiary and secondary amides to amines. The catalyst described herein operates at low loadings using inexpensive 1,1,3,3-tetramethyldisiloxane and allows fast reactions with high yields, selectivities, and turnover numbers. A transient iminium intermediate has been observed for the first time by using mass spectrometry, and the activation of the catalyst and the silane reagent have been studied by using DFT calculations. These fundamental insights support the present and future improvements of Ir^III metallacycles through proper ligand modifications and enable further broad applications of catalysts based on metallacycles.

Full text of DOI:10.1002/cctc.201700400

Accepted Article
Title: Efficient and selective hydrosilylation of secondary and tertiary
amides using an iridium(III) metallacycle catalyst: development
and mechanistic investigation
Authors: Yann Corre, Xavier Trivelli, Frédéric Capet, Jean-Pierre
Djukic, Francine Agbossou-Niedercorn, and Christophe
Michon
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: ChemCatChem 10.1002/cctc.201700400
Link to VoR: http://dx.doi.org/10.1002/cctc.201700400
A Journal of
www.chemcatchem.org

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Efficient and selective hydrosilylation of secondary and tertiary
amides using an iridium(III) metallacycle catalyst: development
and mechanistic investigation.
Yann Corre,[a,b] Xavier Trivelli,[c] Frédéric Capet,[a] Jean-Pierre Djukic,[d] Francine Agbossou-
Niedercorn*[a,b] and Christophe Michon*[a,b]
Dedicated to Professor Jérome Lacour at the occasion of his 50th birthday.
Abstract: In the present study, we show accessible cationic Ir(III)
metallacycles catalyse efficiently the chemoselective
interesting alternative to hydrogenation, provided that
inexpensive and abundant hydrosilanes are used. Because the
reactivity of such reagents and related intermediates is modular
and depends on the substituents of the silicon atoms, the
hydrosilylation of tertiary and secondary amides to amines. The
catalyst described herein operates at low loadings using
inexpensive 1,1,3,3-tetramethyldisiloxane and allows fast
reactions with high yields, selectivities and turnover numbers.
Whereas a transient iminium intermediate has been observed
for the first time by mass spectrometry, the activations of the
catalyst and the silane reagent have been studied by DFT
calculations. These fundamental insights support present and
future improvements of Ir(III) metallacycles through proper
ligand modifications and enable further broad applications of
catalysts based on metallacycles.
hydrosilylation reaction can be
a
highly chemo- and
regioselective reduction method which tolerates various other
reducible functional groups. However, whereas several
organometallic or organic catalysts were shown to hydrosilylate
tertiary and secondary amides,[5] the hydrosilylation of primary
amides remains challenging.[5j,l] Moreover, to the best of our
knowledge, only few publications have reported effective
hydrosilylation of amides by using affordable silanes and highly
active catalysts operating at low loadings.[5b,d,h,i,k] Following our
past studies on hydrosilylation reactions of unsaturated carbon-
carbon and carbon–heteroatom compounds using Ir(III)
metallacycle catalysts,[6] we report herein the first application of
accessible Ir(III) metallacycles to catalyse at low loadings and
high turnover numbers the hydrosilylation of secondary and
tertiary amides using inexpensive 1,1,3,3-tetramethyldisiloxane.
Introduction
Amines are ubiquitous in natural products, building blocks or
targets for fine chemicals, farming-related chemicals and
biologically active compounds.[1] They can be prepared by
numerous synthetic methodologies including reductive
aminations,[1c] alcohol aminations,[1d] hydroaminations[1e] and
transaminations.[1f,g] However, when the preparation of more
functionalised amines is required by pharmacy and fine
chemistry, the reduction of amides is often preferred. Nowadays,
such reaction still involves the sub-stoichiometric use of metal
hydride reagents which are air and moisture sensitive and
produce large amounts of waste.[2] Whereas the reduction of
amides can be performed by hydrogenation using
heterogeneous catalysts under harsh conditions, the use of
homogeneous catalysts allow more active and selective
hydrogenation reactions by operating in milder conditions.[3]
Alternatively, the use of hydrosilanes as reductants is an area of
growing interest for the mild and selective reduction of carboxylic
acid derivatives by using transition metal or Lewis acid
catalysts.[4,5] Indeed, hydrosilylation, which operates without any
high pressure equipment and high temperature, can be an
Results and Discussion
Following our previous investigations on hydrosilylation
reactions,[6] screening of Ir(III) metallacycle precursors,
additives and reaction conditions was performed on the
hydrosilylation of tertiary amide 5a (Table 1). In presence of
2 mol% of trityltetra(pentafluorophenyl)borate (Ph3CBArF20)
and triethylsilane, chromiumtricarbonyl-bound iridacycle
provided a more active catalyst as compared to iridacycle
affording amine 6a in yields of 48 versus 33% (entries 1, 2).
Addition of an NMe2 donating substituent to the chelating 2-
phenyl-pyridine ligand resulted overall in more active
catalysts. Whereas chromiumtricarbonyl-bound iridacycle
led to 6a in a 91% yield, iridacycle allowed the
hydrosilylation of amide 5a in a nearly quantitative yield
(e.g. 98% yield) (entries 3, 4). Hence, considering the effect
of the Cr(CO)3 moiety, a reverse trend was observed
2
1
4
3
compared to iridacycle
1 and 2. The use of [IrCp*Cl2]2 as
pre-catalyst resulted in a less efficient and selective
reaction with the formation of unidentified side-products
(entry 5). Changing the additive to sodium tetrakis[(3,5-
[a]
Dr. Y. Corre, Dr. F. Capet, Dr. F. Agbossou-Niedercorn, Dr. C.
Michon, Univ. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR
8181 - UCCS - Unité de Catalyse et Chimie du Solide, F-59000 Lille,
France. E-mail: christophe.michon@ensc-lille.fr,
francine.agbossou@ensc-lille.fr
trifluoromethyl)phenyl]borate
dimethylanilinium
(NaBArF24)
tetra(pentafluorophenyl)borate
or
N,N-
(Me2PhHNBArF20) led to a significant decrease of yields
(entries 6 and 7). Finally, triethylsilane was replaced by
inexpensive and green hydrosilylation reagents. Whereas
polymethylhydroxysiloxane (PMHS) afforded amine 6a in a
lower yield (65%), we were glad to reach a quantitative
yield using 1,1,3,3-tetramethyldisiloxane (TMDS) reagent
(entries 8, 9).
[b]
[c]
[d]
ENSCL, UCCS-CCM-MOCAH, (Chimie-C7) CS 90108, 59652
Villeneuve d’Ascq Cedex, France
Dr. X. Trivelli, UGSF CNRS, UMR 8576, Université Lille Nord de
France, 59655 Villeneuve d'Ascq Cedex, France.
Dr. J.-P. Djukic, Institut de Chimie de Strasbourg, UMR 7177,
Université de Strasbourg, 4 rue Blaise Pascal, F-67000 Strasbourg,
France.
Interested by the results obtained in the presence of
Supporting information for this article is given via a link at the end of
the document.
pre-catalyst
3 and TMDS, the reaction conditions were
studied in more details (Table 2). We focused first on the
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
activity of our catalytic system by changing the loadings of
anisole resulted in a 73% yield for amine 6a with a TON of
1460 provided the reaction was run over 24 hours (entries
15-16). Depending on the reaction time, various amounts of
a possible iminium species (vide-infra) could be observed
along with product 6a (entries 15-16).
the pre-catalyst
3 and Ph3CBArF20 additive.
Table 1. Catalytic study.
The scope of the hydrosilylation reaction was studied with
aromatic and aliphatic tertiary amides 5a-v (Table 3). Under the
optimised conditions determined above, the corresponding
amines 6a-v were obtained generally in high to quantitative
yields. Substrates 5a-e bearing only aromatic substituents were
reduced smoothly and the desired products were readily
recovered in 0.5 to 2 hours of reaction (entries 1-5). The
hydrosilylation reaction was effectively working with heterocyclic
substrates 5f-g (entries 6, 7) but compounds 5h-k (entries 8-11)
bearing alkyl cycles on whichever side of the amide required 2 to
8 hours to be reduced in good yields. If the reaction was
apparently not limited by some alkyl substituents (reagents 5h,l
– entries 8, 12), the hydrosilylation of tertiary amides 5m-o
bearing either bulky isopropyl groups or just an ethyl fragment
required up to 48 hours to reach average to good yields (entries
13-15). The same trend was also observed for aromatic amides
5p and 5r bearing a fluoro or a methoxy substituent in ortho-
position (entries 16, 18). Moreover, the developed catalyst
allowed highly chemoselective hydrosilylation reactions
tolerating other reducible functional groups like nitro, cyano and
ester. If substrates 5u-v functionalised by an ester or a nitrile
group in para-position required more time to react completely
(entries 21, 22), other para-substituted derivatives 5q and 5s-t
could be reduced smoothly within 2 to 8 hours in high yields
(entries 17, 19-20).
Entry
pre-cat.
Additive
Silane
Yield
(%)[a]
[b]
1
2
3
4
5
6
7
8
9
1
Ph3CBArF20
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
Et3SiH
TMDS[f]
PMHS[g]
33
48
2
Ph3CBArF20
Ph3CBArF20
Ph3CBArF20
Ph3CBArF20
3
98
4
91
[IrCp*Cl2]2
68[c]
37
[d]
3
3
3
3
NaBArF24
[e]
The reaction scope was further studied with the
hydrosilylation of the less reactive secondary amides 7a-t (Table
4).[5d,5m,5o] To our delight, high yields of the corresponding
amines 8a-t could be obtained at 100°C with 2 equivalents of
TMDS through a simple increase of loadings of pre-catalyst 3
and trityl tetra(pentafluorophenyl)borate additive to respectively
0.5 mol% and 1 mol%. Similar to the reduction of tertiary amides
5a-v, aromatic secondary amides 7a-b and 7d-i were quickly
hydrosilylated with high to quantitative yields (entries 1, 2, 4-9).
As the allyl substituent of substrate 7d was not reduced, we
confirmed again the high chemoselectivity of the catalyst
developed here (entry 4). The hydrosilylation of secondary
amide 7c bearing a bulky tbutyl group needed 24 hours in order
to reach an average yield (entry 3). A decrease of reactivity was
also observed for aromatic amide 7j bearing an ethyl substituent
in ortho-position, 8 hours being required to obtain a good yield
(entry 10). Other ortho- or para-substituted derivatives 7k-n
could be reduced readily, like O-heterocyclic substrate 7o
(entries 11-15). If the hydrosilylation of aliphatic secondary
amides 7p and 7t was also straightforward (entries 16, 20), the
reaction of sterically hindered substrates 7q-s bearing
cyclohexyl or isopropyl groups required 24 to 48 hours to reach
average to good yields (entries 17-19).
Me2PhHNBArF20
76
Ph3CBArF20
100
65[h]
Ph3CBArF20
[a] yield measured by GC; TCE: 1,1’,2,2’-tetrachloroethane. [b] Ph3CBArF20
trityl tetra(pentafluorophenyl)borate. [c] with unidentified side-products. [d]
NaBArF24 sodium tetrakis[(3,5-trifluoromethyl)phenyl]borate. [e]
Me2PhHNBArF20 N,N-Dimethylanilinium tetra(pentafluorophenyl)borate. [f]
TMDS: 1,1,3,3-tetramethyldisiloxane. [g] PMHS: Polymethylhydroxysiloxane.
[h] 91% yield after 45 h reaction.
:
:
:
A control experiment proved the additive was not catalysing
itself the hydrosilylation of amide 5a (entry 1). In the
presence of 1 mol% of pre-catalyst
3, quantitative yields
were obtained at 100 and 80°C (entries 2, 3) but a reaction
at 60°C resulted in a significant decrease of the yield (entry
4). When the pre-catalyst loading was reduced from 1 to 0.5
mol%, the yield was still quantitative at 100°C (entry 5),
while a reaction performed at 80°C resulted in a 58% yield
(entry 6). By operating at 0.1 mol% loading of
3, the yield
was again quantitative within 15h providing a turnover
number (TON) of 1000 (entry 7). We noticed the same
catalytic performances could be maintained decreasing the
amount of TMDS to 2 equivalents, an excess of silane
reagent still being required in order to reach a full
conversion of tertiary amide 5a (entries 8, 9). A further
decrease of the pre-catalyst loading to 0.05 mol% resulted
in a complete reaction within 0.5h with a TON of 2000
(entry 10). Finally, the use of a 0.01 mol% pre-catalyst
loading afforded a quantitative yield with a TON of 10000
provided the reaction was run for 24 hours (entry 11). A part
the decrease of the pre-catalyst loading from 1 to 0.01
mol%, it was worth to note a higher concentration of the
reaction medium had no effect on the yield or the catalyst
activity. Whereas solvents like toluene, 1,4-dioxane or
CPME implied much lower yields (entries 12-14), the use of
In an attempt to improve our catalytic system, we
prepared a new iridacyle by changing the chelating 2-
phenyl-pyridine ligand for a 1-phenyl isoquinoline one while
keeping the NMe2 donating substituent on the phenyl group
(See the Supporting Information). The resulting iridacycle
9
was first applied to the hydrosilylation of the sterically
hindered tertiary amide 5n (Scheme 1). Though this
substrate was reduced in 48 hours using pre-catalyst
3
, the
allowed us to obtain the related tertiary
95% yield in only hours. The
use of iridacycle
9
amine 6n with
a
8
hydrosilylation of the bulky secondary amide 7r was
subsequently studied (Scheme 1).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Table 2. Screening of reaction conditions.
Entry
pre-cat.
(mol%)
TMDS
(eq.)
[5a]
Solvent
T
(°C)
Time
(h)
Yield
TON[b]
(mmol.l-1)
(%)[a]
1[c]
2
0
3
3
3
3
3
3
3
2
1
2
2
2
2
2
2
2
0.077
0.077
0.077
0.077
0.154
0.154
0.770
0.770
0.770
1.540
3.850
1.540
1.540
1.540
1.540
1.540
TCE
TCE[d]
TCE
100
100
80
15
15
15
15
15
15
15
15
15
0.5
24
2
0
-
1
100
100
26
100
3
1
100
4
1
TCE
60
26
5
0.5
0.5
0.1
0.1
0.1
0.05
0.01
0.05
0.05
0.05
0.05
0.05
TCE
100
80
100
58
200
6
TCE
116
7
TCE
100
100
100
100
100
100
100
100
100
100
100
100
57
1000
1000
570
8
TCE
9
TCE
10
11
12
13
14
15
16
TCE
100
100
17
2000
10000
340
TCE
toluene
1,4-dioxane
CPME[e]
anisole
anisole
2
6
120
2
16[f]
45[g]
73[h]
320
0.5
900
24
1460
[a] yield measured by GC; Ph3CBArF20: trityl tetra(pentafluorophenyl)borate. [b] TON, turnover number (mol product / mol catalyst). [c] 2 mol% of
Ph3CBArF20 and no pre-catalyst. [d] TCE: 1,1’,2,2’-tetrachloroethane. [e] CPME: cyclopentylmethylether. [f] 91% yield in 24h. [g] 100% conversion with
55% of possible iminium species (vide-infra) and 45% of 6a. [h] 100% conversion with 27% of possible iminium species (vide-infra) and 73% of 6a
.
Whereas the reaction was performed with a 47% yield in
24h using pre-catalyst , we were glad to obtain a
quantitative yield of secondary amine 8r in only 6 hours
while applying pre-catalyst . In order to understand such a
ligand effect on the catalyst activity, we put all our efforts
into investigating the mechanism of these Ir(III) catalysed
hydrosilylation reactions.
suggested as a key transient intermediate in amide
3
hydrosilylation reactions,[5] it was, to the best of our
knowledge, never evidenced before.
9
Regarding
the
reaction
mechanism,
a
ionic
hydrosilylation[7] pathway could be presumed. However,
some of us had already shown through a combination of
organometallic syntheses and DFT calculations a cohesive
As a preliminary study regarding the possible active
intermediate species involved in the catalytic process and the
mechanism of such hydrosilylation reaction, we performed
analyses by electrospray ionization mass spectrometry (ESI-
MS) in acetonitrile of the crude reaction mixture issued from
the hydrosilylation of tertiary amide 5a at 100°C in 30 minutes
(Table 3, entry 1 and the Supporting Information Figures S1-
S4). The hydrolysed tertiary amine product 6a was clearly
identified and dechlorinated Ir(III) complex 10b could be
deduced through the observed cationic fragments 10b1-b3
combined with BArF20 anion, with or without coordination of
acetonitrile which was the solvent used for the analyses
(Figures S2-S4). Hence, though the reaction mixture was not
quenched, the simple use of wet acetonitrile for the solvation
of the mass analysis sample led to the hydrolysis of the
hydrosilylated reaction product and afforded amine 6a. In
addition, the mass analysis conditions did not allow any
iridium-hydride complex to be observed, probably because of
the possible ionisation of such species during the analysis or
because of reaction with water and air atmosphere. A further
analysis by ESI-MS of the crude mixture resulting from the
hydrosilylation of reagent 5a in anisole allowed us to observe
hydridoiridium(III)
→ silylium donor−acceptor complex could
exist.[8] Hence, we assumed our reaction pathway could
differentiate from the others[4,7] by the activation mode of the
silane. Thermochemical data inferred from singlet state gas
phase DFT-D calculation (ZORA-PBE-D3(BJ)/all electron
TZP basis set) provided a solid basis for establishing a
reasonable mechanistic scheme (Scheme 2) of the therein
reported catalysis particularly for its initiation stage. For
convenience we extended our considerations to precatalysts
1
,
3
and
9
. The computed thermochemistry clearly suggested
to yield
that the necessary dechlorination of precatalyst
3
putative electron deficient cationic intermediate 10b results
from a reaction with an in-situ-produced form of a stabilised
silylium cation generated by the reaction of the tritylium cation
with the silane reagent.[9] For instance, the reaction of Et3SiH
with Ph3C+ is expected to produce triphenylmethane and
cation [Et3Si-H-SiEt3]+ according to Heinekey,[9a] with a Gibbs
enthalpy of –9 kcal/mol according to our DFT-D computations
(Scheme 2, Scheme 3a). Worthy to note, the hypothesis of
the formation of an elusive “naked” trigonal planar silylium
cation[10] by the stoichiometric 1:1 reaction of the tritylium with
Et3SiH was ruled out owing to its high endoergonicity (
+17 kcal/mol). The dechlorination of
by [Et3Si-H-SiEt3]+ to
give intermediary 10a, Et3SiCl and Et3SiH was found
G298=
a mixture of tertiary amine product 6a and iminium species
C
3
(Figure S1, Scheme 2). Though the latter was already
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Table 3. Hydrosilylation of tertiary amides catalysed by iridacycle 3.
Entry
Amide
R1
R2
R3
t
Yield[a]
(h)
(isolated yield %)[b]
1
2
3
5a
5b
5c
Ph
Ph
Ph
Bn
Me
Ph
Bn
Me
Ph
0.5
0.5
2
6a 100 (92)
6b 100 (96)
6c 100 (80)
4
5d
5e
5f
Ph
Ph
Ph
Bn
Me
Me
0.5
0.5
0.5
0.5
8
6d 100 (93)
6e 100 (97)
6f 100 (87)
6g 100 (99)
6h 83 (81)
6i 85 (82)
5
6
Ph
-(CH2)2-O-(CH2)2-
7
5g
5h
5i
1-furane
Cy
Bn
Bn
Bn
Bn
8
9
Ph
-(CH2)5-
-(CH2)5-
8
10
11
12
13
14
15
16
17
18
19
20
21
22
5j
H
6
6j 100 (99)
6k 100 (45)
6l 100 (92)
6m 49[c] (39)
6n 78 (60)
6o 81 (65)
6p 79 (69)
6q 100 (99)
6r 50 (43)
5k
5l
-(CH2)4-
Bn
Bn
Bn
Bn
iPr
Bn
Bn
Bn
Bn
Bn
Bn
Bn
2
PhCH2CH2-
Et
Bn
Bn
Bn
iPr
Bn
Bn
Bn
Bn
Bn
Bn
Bn
2
5m
5n
5o
5p
5q
5r
24
48
48
48
8
iPr
Ph
(o-MeO)C6H4-
(p-MeO)C6H4-
(o-F)C6H4-
(p-F)C6H4-
(p-NO2)C6H4-
(p-CN)C6H4-
(p-CO2Me)C6H4-
24
4
5s
5t
6s 99 (92)
6t 93 (77)
2
5u
5v
20
24
6u 98 (91)
6v 100 (90)
[a] Yield measured by GC. [b] Isolated yield after flash chromatography.
[c] 6m along with 17% of dibenzylamine resulting from the hydrolysis of the related iminium intermediate.
in this case to be largely exoergonic with a Gibbs enthalpy of
-31 kcal/mol (Scheme 3b).
reaction implying the formation of molecular complexes of the
trimethylsilylium cation;[10a] silylicity characterises in analogy
with hydricity[11] the capability of a given silyl-metal complex
to release a silylium group.
The alternative dechlorination of
3
by the tritylium was found
to be endoergonic with a Gibbs enthalpy

G298 of ca. +6
The isodesmic reaction depicted in Scheme 3d shows the
hypothetical transfer of silylium from 11b and 11c respectively
towards 12a, which is nearly isoergonic for the former pyridyl
derivative and endoergonic by ca. 4 kcal/mol for the latter
quinoline derivative. This higher affinity of the silylium group for
the hydrido-iridium unit in 11c cannot be exclusively assigned to
the strong donor effect of the N,N-dimethylamino group because
such an effect is not observed with 11b. We speculate that the
non-coplanarity of the aryl fragments in the quinoline derivative
11c might restrict the extent of the mesomeric electron
delocalisation operated by the amino group towards the
quinoline group, allowing a slightly larger charge density to
concentrate at the metal centre than in the pyridyl analogue 11b,
which contains a planar iridacycle. A Ziegler-Rauk energy
decomposition analysis (EDA)[12] was performed on 11a, 11b
and 11c to probe the intrinsic bonding energy of the silylium
Et3Si+ cation, or intrinsic silylicity, with the hydridoiridium
fragment; it shows clearly that the interaction energy Eint of
kcal/mol and was therefore not considered further. The
reaction of iridium catalyst 10b with Et3SiH produces the
silane–iridium adduct 11b with a Gibbs enthapy of -13
kcal/mol (Scheme 3c).[10,11] Worthy to note here the
production of 11b is by ca. 5 kcal/mol less exoergonic, that is
less favorable than those of the adducts resulting from
, i.e 11a and 11c. Still, the silylium fragment has a greater
affinity for the hydrido-iridium unit in 11c than in 11a and 11b
1 and
9
,
the latter two displaying identical relative silylicity (Scheme
3d). It must be pointed out here that ongoing research is
focused on the mechanism of formation of intermediates 11
from 10 that will be disclosed elsewhere: preliminary results
from DFT investigations point to a barrier-less process
corresponding to a concerted hydride transfer and silylium
trapping from Et3Si-H.
The term silylicity proposed here is more general than the
specific TMSA (Trimethyl silyl affinity) introduced by Villinger
et al. for computing thermodynamic parameters of model
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Table 4. Hydrosilylation of secondary amides catalysed by iridacycle 3.
Entry
Amide
R1
R2
t
Yield[a]
(h)
(isolated yield %)[b]
1
2
3
7a
7b
7c
Ph
Ph
Ph
Bn
Me
tBu
0.5
0.5
24
8a 100 (quant.)
8b 100 (85)
8c 49 (46)
4
5
7d
7e
7f
Ph
Ph
allyl
0.5
0.5
2
8d 100 (82)
8e 100 (86)
8f 47 (45)
-CH2-1-thiophene
6
Ph
cyclohexyl
7
7g
7h
7i
Ph
Ph
0.5
0.5
0.5
8
8g 98 (92)
8h 100 (96)
8i 96 (75)
8
Ph
(p-MeO)C6H4
9
Ph
(p-F)C6H4-
10
11
12
13
14
15
16
17
18
19
20
7j
Ph
(o-Et)C6H4-
8j 83 (82)
7k
7l
(o-MeO)C6H4-
(p-MeO)C6H4-
(o-F)C6H4-
(p-F)C6H4-
1-furane
PhCH2CH2-
cyclohexyl-
iPr
Bn
Bn
2
8k 94 (76)
8l 96 (96)
0.5
0.5
0.5
0.5
0.5
24
24
48
2
7m
7n
7o
7p
7q
7r
Bn
8m 99 (86)
8n 100 (quant).
8o 100 (79)
8p 100 (quant.)
8q 68 (44)
8r 47 (40)
Bn
Bn
Bn
Bn
cyclohexyl
Bn
7s
7t
iPr
8s 81 (65)
8t 100 (76)
-(CH2)4-
[a] Yield measured by GC. [b] Isolated yield after flash chromatography.
Scheme 1. Improved catalyst for the reduction of tertiary and secondary amides.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Scheme 2. Proposed catalytic cycle for the reduction of tertiary and secondary amides.
Scheme 3. Thermochemical data computed (298.15 K) at the ZORA-PBE-
D3(BJ)/all electron TZP level with gas phase ground state molecules devoid of
counter anion BArF20
and in 11c (Eint= -127 kcal/mol) than in 11a (Eint= -119
kcal/mol). Consistently the total natural charge (Natural
Population Analysis) borne by the Et3Si group in 11a (q=
+0.29) is slightly higher than that in 11b (q= +0.25) for
example.
.
In the framework of the mechanism depicted in
Scheme 2, adducts 11a-c are sources of hydrido-iridium
intermediates 12a-c and silylium cation R3Si+.[9,10,13] The
latter may activate the carbonyl group of amide substrate
5a and generate silyloxy carbonium species
with a first equivalent of the iridium hydride complex 12
affords silyl hemiacetal along with the cationic iridium
A. Reaction
B
complex 10. At this stage, elimination of a silyloxide
fragment may be helped by any electrophilic species
present in the medium to lead to the iminium intermediate
C. The latter can then react with a second equivalent of the
iridium hydride complex 12 and affords the amine product
6a along with the cationic iridium catalytic species 10
(Scheme 2). As mentioned in Table 2 (entries 8, 9), 2
equivalents of TMDS were added in order to get a
quantitative hydrosilylation of amide 5a. At the end of
reaction, we could isolate the silicon residue from the
resulting crude mixture. Interestingly, a 1H NMR analysis
revealed the presence of a hydride silicon species and
suggested only one hydride from the two of TMDS reagent
may participate to the reaction (see the Supporting
Information Figures S5, S6). Further characterisation by
ESI-MS of such a silicon residue formed at the end of
reaction showed several silicon compounds were formed
during the catalysis (Figure S7).
the silylium group with the neutral hydrido-iridium fragment
is by ca. 8 kcal/mol stronger in 11b Eint= -127 kcal/mol)
(
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Point 1: all three pre-catalysts
dechlorinated readily, with a somewhat larger exoergon-
icity for and (Scheme 3b).
1, 3 and 9 can be
3
9
Point 2: 10a gives less exo-ergonically a silane-iridium
adduct than 10c (Scheme 3c); this rather counter-intuitive
result can be explained by the major inter-annular tensions
that exist within 10c and which are somewhat released
upon return of the Ir(III) center to a hepta- or hexa-
coordinate geometry like in 11c and 12c respectively
(Figure 1). According to the X-ray structure determination, it
is worth to note inter-annular tensions are also observed for
chlorinated pre-catalyst
Point 3: 11a and 11b have nearly the same relative
silylicity ( G= kcal/mol, Scheme 3d) whereas 11c
9 (Figure S8).

0
displays a lower one (
Figure 1).
G= +4 kcal/mol, Scheme 3d and
Point 4: both 12a and 12c display similar hydricity
(Scheme 3e and Figure 1).
It seems therefore tempting to state that the
outstanding performance of
9 could be related to the easier
formation of silane-iridium adduct 11c combined with its
relatively lower silylicity and to a relatively higher hydricity of
its related hydrido complex 12c
. A high hydricity is
a)
intuitively related to a faster reduction of an electrophile of
reference. A slightly lower relative intrinsic silylicity might
sensibly contribute in decreasing the weight of the pathway
detrimental to the catalysis that entails the formation of
catalytically inactive -hydrido bridged bis-iridium species. It
was shown that such species are formed almost
unavoidably in rather large amounts when a cationic
solvato-iridacycle derived from 2-phenylpyridine is treated
with Et3SiH at room temperature.[8]
Conclusions
To summarise, we have described a highly efficient
catalytic process for the hydrosilylation of tertiary and
secondary amides. The combination of an accessible Ir(III)
b)
metallacycle
complex
and
trityltet-
Figure 1. a) Gas phase singlet state geometry of 11c optimised at the
ZORA-PBE-D3(BJ)/all electron TZP level (interatomic distances printed
in green are expressed in Å); b) Similarly to 11b the NCI plot[14] of 11c
ra(pentafluorophenyl)borate in the presence of 1,1,3,3-
tetramethyldisiloxane allowed, at low loadings, the
chemoselective reduction of a large array of amide sub-
strates in high yields and turnover numbers tolerating a
wide range of substituents. According DFT calculations,
depicts
computed model were covalent and donor acceptor interactions are
identical to those already pointed out for 11b[14]
ADFview2013 plot of
non-covalent interaction (NCI) regions (ZORA-PBE-D3(BJ)/ae-TZP)
materialised by reduced density gradient isosurfaces (cut-off value s =
a rather non-covalent attractive H…Si interaction in this
:
silane-iridium adduct is
a
source of hydrido-iridium
0.02 a.u.,
 = 0.05 a.u.) colored according to the sign of the signed
intermediate and silylium cation. The latter activates the
carbonyl group of the amide substrate and generates a
silyloxy carbonium species. Reaction with a first equivalent
of the iridium hydride complex affords silyl hemiacetal along
with the cationic catalytic species. The further elimination of
silyloxide fragment leads to an iminium intermediate which
has been observed my mass spectrometry for the first time.
Reaction with a second equivalent of the iridium hydride
complex affords the amine product along with the cationic
catalytic species. On the whole, the work described herein
demonstrates the reactivity of Ir(III) metallacycles can be
improved through proper ligand modifications to allow the
straightforward catalytic hydrosilylation of challenging
amide substrates. Such developments foresee clearly
future broad and significant applications of catalysts based
on metallacycles.
density 2, that is red (attractive) and blue (Pauli repulsive or non-
bonded van der Waals) isosurfaces are associated with negatively and
positively signed terms respectively.
Addressing the difference in reactivity of the two ami-
no-substituted precatalysts
difficult task in such a catalysis where a great number of
species like substrate and cocatalysts coexist.
3 and 9 as compared to 1 is a
Nevertheless, even though further efforts will be needed in
the future, thermochemistry of possible reactions can be
calculated by DFT computations and offer sufficient clues to
attempt a rationalisation. At this stage, assuming the
reactions to be mostly driven by favourable initiation and
catalytic event under thermodynamic control, we can only
elaborate on four points of crucial importance that are the
readiness of: 1) the precatalyst’s dechlorination, 2) the
reaction of silane with transient 10 to give 11, 3) silylium
abstraction from 11, and 4) the hydride transfer from the
hydrido-Ir transient.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Experimental Section
Acknowledgements
The University of Lille 1 and Région Hauts-de-France
are acknowledged for PhD fellowship (Y. C.). The CNRS,
the Chevreul Institute (FR 2638), the Ministère de
l’Enseignement Supérieur et de la Recherche, the LABEX
Chimie des Systèmes Complexes (Strasbourg), the Région
Hauts-de-France and the FEDER are acknowledged for
supporting and funding partially this work. Mrs Catherine
Méliet (UCCS) is thanked for elemental analyses. Mrs
Céline Delabre (UCCS) is thanked for GC and GC-MS
analyses. Mrs Nathalie Duhal and Céline Lenglart (CUMA,
Univ. of Lille) are thanked for HRMS analyses. Ms Thi
Nguyet Anh Ho (ENSCL) and Ms Catrina Larisa Georgiana
(Erasmus exchange between ENSCL, France and
University of Craiova, Romania) are thanked for their help
with experimental work.
General Procedure for the catalysis
In a Schlenk tube, amide reagent (1.54 mmol, 1 eq.)
and iridium (III) catalyst (0.05 mol%, 0.0005 eq.) are
introduced. BArF salt (0.1 mol%) is then added in a
glovebox. Under nitrogen, 1 mL of TCE and TMDS (2 eq.)
are transferred by syringe and the reaction mixture is
heated at 100 °C under stirring for 2 hours (the Schlenk
tube being closed under N2). Afterwards, the solvent is
evaporated under vacuum (using a Schlenk line) and the
reaction mixture is hydrolysed following method A or B.
Method A: acid hydrolysis (examples: hydrosilylation
of amines 5a, 5b, 5e, 5f, 5g). After complete evaporation of
the solvent, the reaction mixture is diluted with diethylether
(4 mL) and HCl 4M (4 mL) is added. The resulting solution
is stirred vigorously during 10 minutes to afford the
protonated amine as a precipitate which was recovered by
filtration and washed with diethylether. The resulting solid
was then disolved in CH2Cl2 and neutralized with an
aqueous solution of NaOH 1M (for tertiary amines) or of
saturated NaHCO3 (for secondary amines) until pH > 7.
After extraction (3 times with CH2Cl2), the organic phase
was dried with MgSO4 and evaporated to afford the desired
amine.
Keywords: iridium • metallacycles • hydrosilylation • amides •
amines.
[1]
a) K. S. Hayes, Appl. Catal. A: Gen. 2001, 221, 187; b) S. A. Lawrence,
Amines: Synthesis, Properties and Applications 34 (Cambridge
University Press, 2004); c) A. F. Abdel-Magid, S. J. Mehrman, Org.
Process Res. Dev. 2006, 10, 971; d) Q. Yang, Q. Wang, Z. Yu, Chem.
Soc. Rev. 2015, 44, 2305; e) L. Huang, M. Arndt, K. Gooßen, H. Heydt,
L. J. Gooßen, Chem. Rev. 2015, 115, 2596; f) M. Breuer, K. Ditrich, T.
Habicher, B. Hauer, M. Keßeler, R. Stürmer, T. Zelinski, Angew. Chem.
2004, 116, 806; Angew. Chem. Int. Ed. 2004, 43, 788; g) M. D. Truppo,
J. D. Rozzell, N. J. Turner, Org. Process Res. Dev. 2010, 14, 234.
a) G. W. Gribble, Chem. Soc. Rev. 1998, 27, 395; (b) J. Seyden-Penne,
Reductions by the Alumino- and Borohydrides in Organic Synthesis,
(2nd ed.; Wiley: New York, 1997).
Method B: basic hydrolysis (examples: hydrosilylation
of amines 5c, 5d). After complete evaporation of the
solvent, the reaction mixture is diluted with methanol (4mL)
and NaOH 1M (4 mL) is added. After extraction (3 times
with Et2O), the organic phase was dried with MgSO4 and
evaporated to afford the desired amine which was further
purified by flash chromatography using mixtures of
petroleum ether and ethyl acetate with 5% NEt3.
[2]
[3]
a) Catalysed hydrogenation of amides to amines: a) S. Das, S. Zhou, D.
Addis, S. Enthaler, K. Junge, M. Beller, Top. Catal. 2010, 53, 979; b) P.
A. Dub, T. Ikariya, ACS Catal. 2012, 2, 1718; c) S. Werkmeister, K.
Junge, M. Beller, Org. Process Res. Dev. 2014, 18, 289 and references
therein; d) T. vom Stein, M. Meuresch, D. Limper, M. Schmitz, M.
Hꢀlscher, J. Coetzee, D. J. Cole-Hamilton, J. Klankermayer, W. Leitner,
J. Am. Chem. Soc. 2014, 136, 13217; e) E. Balaraman, D. Milstein,
Topics in Organomet. Chem. 2015, 48, 19; f) J. M. John, R. Loorthuraja,
E. Antoniuk, S. H. Bergens, Catal. Sci. Technol. 2015, 5, 1181; g) J. R.
Cabrero-Antonino, E. Alberico, K. Junge, H. Junge, M. Beller, Chem.
Sci. 2016, 7, 3432; h) J. A. Garg, S. Chakraborty, Y. Ben-David, D.
Milstein, Chem. Commun. 2016, 52, 5285; i) J. R. Cabrero-Antonino, E.
Alberico, H. J. Drexler, W. Baumann, K. Junge, H. Junge, M. Beller,
ACS Catalysis 2016, 6, 47; j) M. L. Yuan, J. H. Xie, S. F. Zhu, Q. L.
Zhou, ACS Catalysis 2016, 6, 3665; k) M. L. Yuan, J. H. Xie, Q. L. Zhou,
ChemCatChem 2016, 8, 3036.
Computational details
Computations were performed with methods of Density
Functional Theory, i.e the PBE[15] GGA functional implemented
in the Amsterdam Density Functional package (ADF2013[16]
version) and augmented with Grimme’s DFT-D3(BJ)[17]
implementation of dispersion with
a Becke-Johnson (BJ)
damping function. Scalar relativistic corrections with the Zeroth
Order Regular Approximation[18] were applied with ad hoc all-
electron (AE) basis sets consisting of polarised triple- (TZP)
Slater type orbitals. Geometry optimisation by energy gradient
minimisation was carried out in all cases with a numerical grid
accuracy comprised between 4.5 and 8, an energy gradient
convergence criterion of 10-3 au and
a very tight SCF
convergence criterion. Vibrational modes were computed to
verify that the optimised geometries were related to energy
minima not considering residual modes comprised between 0
and 100i cm-1. Fragment interaction energies were extracted
from a conventional Energy Decomposition Analysis according
to the Ziegler-Rauk method implemented in ADF.[11] By
extension, intinsic hydricity and silylicity have been determined
by considering the disruption of a hydrido-metal and silyl-metal
intermediate by confering to the hydrogen a negative charge
(hydride) and to the silyl fragment a positive charge (silylium in
its prepared geometry). This arbitrary choice of fragmentation,
although artificial, gives an intuitive direct access to the intrinsic
affinity of the metal center for those ligands that are suspected
to play a central role in the catalysis depicted here according to
a recent report.[8]
[4]
a) B. Marciniec, H. Maciejewski, C. Pietraszuk, P. Pawluć, in
Hydrosilylation: A Comprehensive Review on Recent Advances, (Ed. B.
Marciniec, Springer, Heidelberg, 2009); b) D. Addis, S. Das, K. Junge,
M. Beller, Angew. Chem. 2011, 123, 6128; Angew. Chem. Int. Ed. 2011,
50, 6004; c) K. Revunova, G. I. Nikonov, Dalton Trans. 2015, 44, 840;
d) J. Pesti, G. L. Larson, Org. Process Res. Dev. 2016, 20, 1164; e) M.
Oestreich, Angew. Chem. 2016, 128, 504, Angew. Chem. Int. Ed. 2016,
55, 494; f) M. C. Lipke, A. L. Liberman-Martin, T. D. Tilley, Angew.
Chem. 2017, 129, 2298; Angew. Chem. Int. Ed. 2017, 56, 2260.
Catalysed hydrosilylation of amides to amines: a) Y. Motoyama, K.
Mitsui, T. Ishida, H. Nagashima, J. Am. Chem. Soc. 2005, 127, 13150;
b) S. Hanada, E. Tsutsumi, Y. Motoyama, H. Nagashima, J. Am. Chem.
Soc. 2009, 131, 15032; c) Y. Sunada, H. Kawakami, T. Imaoka, Y.
Motoyama, H. Nagashima, Angew. Chem. 2009, 121, 9675; Angew.
Chem. Int. Ed. 2009, 48, 9511; d) Y. Motoyama, M. Aoki, N. Takaoka,
[5]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
R. Aoto, H. Nagashima, Chem. Commun. 2009, 45, 1574; e) S. Zhou, K.
G. I. Nikonov, J. Am. Chem. Soc. 2011, 133, 6487; g) J. Yang, P. S.
White, M. Brookhart, J. Am. Chem. Soc. 2008, 130, 17509; h) R. M.
Bullock, Chem. Eur. J. 2004, 10, 2366.
Junge, D. Addis, S. Das, M. Beller, Angew. Chem. 2009, 121, 9671,
Angew. Chem. Int. Ed. 2009, 48, 9507; f) S. Das, D. Addis, S. Zhou, K.
Junge, M. Beller, J. Am. Chem. Soc. 2010, 132, 1770; g) S. Das, D.
Addis, K. Junge, M. Beller, Chem. Eur. J. 2011, 17, 12186; h) S. Park,
M. Brookhart, J. Am. Chem. Soc. 2012, 134, 640; i) C. Cheng, M.
Brookhart, J. Am. Chem. Soc. 2012, 134, 11304; j) Y. Li, J. A. Molina
de La Torre, K. Grabow, U. Bentrup, K. Junge, S. Zhou, A. Brückner, M.
Beller, Angew. Chem. 2013, 125, 11791; Angew. Chem. Int. Ed. 2013,
52, 11577; k) T. Dombray, C. Helleu, C. Darcel, J.-B. Sortais, Adv.
Synth. Catal. 2013, 355, 3358; l) E. Blondiaux, T. Cantat, Chem.
Commun. 2014, 50, 9349; m) O. O. Kovalenko, A. Volkov, H. Adolfsson,
Org. Lett. 2015, 17, 446; n) S. Das, Y. Li, L.-Q. Lu, K. Junge, M. Beller,
Chem. Eur. J. 2016, 22, 7050;o) B. Li, J.-B. Sortais, C. Darcel, RSC
Advances 2016, 6, 57603; p) F. Tinnis, A. Volkov, T. Slagbrand, H.
Adolfsson, Angew. Chem. 2016, 128, 4638; Angew. Chem. Int. Ed.
2016, 55, 4562; q) A. Augurusa, M. Mehta, M. Perez, J. Zhu, D. W.
Stephan, Chem. Commun. 2016, 52, 12195.
[8]
[9]
M. Hamdaoui, M. Ney, V. Sarda, L. Karmazin, C. Bailly, N. Sieffert, S.
Dohm, A. Hansen, S. Grimme. J.-P. Djukic, Organometallics 2016, 35,
2207.
a) S. J. Connelly, W. Kaminsky, D. M. Heinekey, Organometallics 2013,
32, 7478; b) M. Nava, C. A. Reed, Organometallics 2011, 30, 4798; c)
S. P. Hoffmann, T. Kato, F. S. Tham, C. A. Reed, Chem. Commun.
2006, 767.
[10] a) M. F. Ibad, P. Langer, A. Schulz, A. Villinger, J. Am. Chem. Soc.
2011, 133, 21016; b) C. A. Reed, Acc. Chem. Res. 1998, 31, 325.
[11] E. S. Wiedner, M. B. Chambers, C. L. Pitman, R. M. Bullock, A. J.-M.
Miller, A. M. Appel, Chem. Rev. 2016, 116, 8655.
[12] T. Ziegler, A. Rauk, Inorg. Chem. 1979, 18, 1755.
[13] a) C. Scheeren, F. Maasarani, A. Hijazi, J.-P. Djukic, M. Pfeffer, S. D.
Zaric, X.-F. LeGoff, L. Ricard, Organometallics 2007, 26, 3336; b) Y. Hu,
L. Li, A. P. Shaw, J. R. Norton, W. Sattler, Y. Rong, Organometallics
2012, 31, 5058.
[6]
a) W. Iali, F. La Paglia, X.-F. Le Goff, D. Sredojevic, M. Pfeffer, J.-P.
Djukic, Chem. Commun. 2012, 48, 10310; b) Y. Corre, W. Iali, M.
Hamdaoui, X. Trivelli, J.-P. Djukic, F. Agbossou-Niedercorn, C. Michon,
Catal. Sci. Technol. 2015, 5, 1452; c) C. Michon, K. MacIntyre, Y. Corre,
F. Agbossou-Niedercorn, ChemCatChem 2016, 8, 1755; d) Y. Corre, C.
Werlé, L. Brelot-Karmazin, J.-P. Djukic, F. Agbossou-Niedercorn, C.
Michon, J. Mol. Catal. A: Chem. 2016, 423, 256; e) Y. Corre, V. Rysak,
F. Capet, J.-P. Djukic, F. Agbossou-Niedercorn, C. Michon, Chem. Eur.
J. 2016, 22, 14036.
[14] a) J. Contreras-Garcia, E. R. Johnson, S. Keinan, R. Chaudret, J. P.
Piquemal, D. N. Beratan, W. Yang, J. Chem. Theor. Comput. 2011, 7,
625; b) E. Johnson, S. Keinan, P. Mori-Sanchez, J. Contreras-Garcia, A.
J. Cohen, W. Yang, J. Am. Chem. Soc. 2010, 132, 6498.
[15] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[16] K. Vanka, M. S. W. Chan, C. C. Pye, T. Ziegler, Organometallics 2000,
19, 1841.
[17] a) S. Grimme, S. Ehrlich, L. Goerigk, J. Comp. Chem. 2011, 32, 1456;
b) S. Grimme, Wiley Interdisciplinary Reviews: Computational
Molecular Science 2011, 1, 211.
[7]
a) M. Oestreich, J. Hermeke, J. Mohr, Chem. Soc. Rev. 2015, 44, 2202;
b) M. Iglesias, F. J. Fernꢁndez-Alvarez, L. A. Oro, ChemCatChem 2014,
6, 2486; c) T. T. Metsänen, P. Hrobárik, H. F. T. Klare, M. Kaupp, M.
Oestreich, J. Am. Chem. Soc. 2014, 136, 6912; d) W. Wang, P. Gu, Y.
Wang, H. Wei, Organometallics 2014, 33, 847; e) M. Iglesias, P. J.
Sanz-Miguel, V. Polo, F. J. Fernꢁndez-Alvarez, J. J. Pérez-Torrente, L.
A. Oro, Chem. Eur. J. 2013, 19, 17559; f) O. G. Shirobokov, L. Kuzmina,
[18] a) E. van Lenthe, E.-J. Baerends, J. G. Snijders, J. Chem. Phys. 1993,
99, 4597; b) E. van Lenthe, E.-J. Baerends, J. G. Snijders, J. Chem.
Phys. 1994, 101, 9783; c) E. van Lenthe, A. Ehlers, E.-J. Baerends, J.
Chem. Phys. 1999, 110, 8943.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201700400
ChemCatChem
FULL PAPER
Straightforward and selective reduction of amides into amines
FULL PAPER
Cationic iridium (III) metallacycle
complexes are highly active catalysts
for the chemoselective reduction of
amides to amines through
Yann Corre, Xavier Trivelli, Frédéric
Capet, Jean-Pierre Djukic, Francine
Agbossou-Niedercorn,* Christophe
Michon*
hydrosilylation reaction. A transient
iminium intermediate has been
observed for the first time by mass
spectrometry. The activations of the
catalyst and the silane reagent have
been studied by DFT calculations.
Page No. – Page No.
Efficient and selective hydrosilylation
of secondary and tertiary amides
using an iridium(III) metallacycle
catalyst: development and
mechanistic investigation.
This article is protected by copyright. All rights reserved.