Iridium-PPh3 Catalysts for Conversion of Amides to Enamines

Article abstract of DOI:10.1021/acs.organomet.8b00835

Studies on the deactivation mechanism of the reaction of N,N-dialkylamides with TMDS catalyzed by Vaska's complex, IrCl(CO)(PPh₃)₂ (1a), triggered the discovery of highly active Ir-PPh₃ catalysts: Photochemically activated

Full text of DOI:10.1021/acs.organomet.8b00835

Article
pubs.acs.org/Organometallics
Cite This: Organometallics XXXX, XXX, XXX−XXX
Iridium-PPh3 Catalysts for Conversion of Amides to Enamines
Yuta Une,† Atsushi Tahara,‡ Yasumitsu Miyamoto,† Yusuke Sunada,§ and Hideo Nagashima*,†,‡
‡
†Interdisciplinary Graduate School of Engineering Science and Institute for Materials Chemistry and Engineering, Kyushu
University, Kasugakoen 6-1, Kasuga, Fukuoka 816-8580, Japan
§Institute of Industrial Science, The University of Tokyo, Komaba 4-6-1, Meguro, Tokyo 153-8505, Japan
S
* Supporting Information
ABSTRACT: Studies on the deactivation mechanism of the
reaction of N,N-dialkylamides with TMDS catalyzed by
Vaska’s complex, IrCl(CO)(PPh3)2 (1a), triggered the
discovery of highly active Ir-PPh3 catalysts: photochemically
activated 1a and thermally activated IrCl(PPh3)3 (8). Both
catalysts showed excellent activity toward the selective
conversion of a variety of N,N-dialkyl-, N-alkyl-N-aryl-, and
N,N-diarylamides to the corresponding enamines with low catalyst loadings. The 14-electron species “ClIr(PPh3)2”, which is
stabilized by solvents or reactants in the actual catalytic reactions, could be involved in the catalysis, which produces
“HIr(PPh3)2” and “SiIr(PPh3)2” (Si = Me2HSiOMe2Si−) species in the catalytic cycle. An in situ generation method for the
“ClIrL2” species was established by simply mixing [IrCl(η4-COD)]2 with PPh3 or other phosphorus ligands, which realized the
facile large-scale syntheses of enamines.
available, storable under aerobic conditions, and easy to
handle. Among them, TMDS is especially useful and is
inexpensive enough to be applied to industrial reduction
processes.8
INTRODUCTION
■
Hydrosilanes are stable metal hydrides that are useful for
carbonyl reduction in the presence of Brønsted acids, Lewis
acids, fluoride anion, and transition-metal catalysts.1 Recent
progress in catalyst design to achieve high catalytic efficiency
has led to the discovery of new catalysts containing not only
conventional precious metals but also base metals, attracting
the attention of scientists due to their environmentally benign
nature.2 While the initial stage of studies on metal-catalyzed
reduction with hydrosilanes was limited to the hydrosilylation
of aldehydes and ketones, recent studies have focused on
hydrosilane reduction of carboxylic acids or esters to silyl
ethers as well as amides to amines.3 It is known that the
reduction of carboxylic acid derivatives is more difficult than
that of aldehydes and ketones with conventional alumino- and
borohydrides.1,2
One of the keys to achieving the efficient hydrosilane
reduction of carboxylic acid derivatives is the contribution of
hydrosilane structures to increase the reduction rates. As
summarized by our accounts and perspective,4 we have
investigated the high reactivity of bifunctional hydrosilanes,
in which two Si−H groups are located in close proximity.
Bifunctional hydrosilanes such as Me2HSi(CH2)2SiHMe2
(BDMSE), Me2 HSiOSiHMe2 (TMDS), Me3 Si-
(OSiHMe)nOSiMe3 (PMHS), and 1,2-dimethylsilylbenzene
(BDSB) are capable of forming disilametallacyclic intermedi-
ates by a reaction with metal complexes and take part in the
efficient reduction of carboxylic acid derivatives: in particular,
the conversion of amides to amines. A combination of these
bifunctional hydrosilanes with ruthenium,5 platinum,6 and
iron7 catalysts realized the reduction of secondary and tertiary
amides to the corresponding secondary and tertiary amines,
respectively. These bifunctional hydrosilanes are commercially
Among the metal-catalyzed hydrosilane reductions of
amides, those catalyzed by Vaska-type iridium complexes,
IrCl(CO)L2, where L is a phosphorus ligand, are unique in not
affording amines but selectively providing enamines through
silylhemiaminal intermediates.9a Among hydrosilanes including
the bifunctional hydrosilanes described above, only TMDS and
PMHS successfully promoted the conversion of amides to
enamines. No reaction occurred with other bifunctional
hydrosilanes such as BDSE and BDMSE or monofunctional
primary, secondary, and tertiary hydrosilanes such as PhSiH3,
Ph2SiH2, and PhMe2SiH.
Vaska-type iridium complexes are surprisingly efficient for
the conversion of amides to enamines with TMDS. As shown
in the upper equation in Scheme 1, N,N-dialkylamides 2
underwent the reaction with TMDS by catalysis of IrCl(CO)-
(PPh3)2 (1a) at room temperature. The corresponding
enamines 3 were formed as a single product in quantitative
yields within 1 h. In an extreme case, the TON and TOF (h−1)
reached 8000 and 16000, respectively. The reaction is
considered to proceed through a silylhemiaminal intermediate,
which is not detectable in the reaction of 2. π-Conjugated
enamines are attractive donor molecules in materials science.10
The preparation of a series of π-conjugated enamines from
N,N-diaryl-2-arylacetamides 4 and TMDS has not been very
successful by catalysis of 1a. However, they were efficiently
hydrosilylated to the corresponding silylhemiaminals 5 by the
Received: November 14, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
molecule of TMDS to the resulting IrH(CO)(PPh3)2 was
followed by elimination of H2 to form 7. The 1H NMR
spectrum of 7 showed a characteristic triplet at δ −7.66 in
Scheme 1. Iridium-Catalyzed Hydrosilane Reduction of
Amides, Leading to Efficient Synthesis of Aldenamines
C6D6 due to Ir−H. The Ir−H signal was coupled with two 31
P
nuclei at the cis position with JP−H = 16.5 Hz. A 31P{1H}
resonance appeared as a singlet at 1.95 ppm.
When the reaction of PhCH2CONEt2 (2a) with TMDS was
carried out in toluene at room temperature in the presence of
0.05 mol % of 1a, the reaction was complete in 1 h to give
PhCHCHNEt2 (3a) quantitatively (entry 1, Table 1). The
reaction performed with lower catalyst loading (0.001 mol %;
entry 2) was terminated at 25% conversion of 2a, as seen in the
reaction profile in Figure 1. Analysis of the iridium species in
solution after the reaction revealed the existence of 7 as a
catalysis of IrCl(CO)[P(OC6F5)3]2 (1b).9b The silylhemiami-
nals 5 were converted to π-conjugated enamines 6 either by
thermal treatment or by the action of acid. The catalytic
activity of the iridium complex 1b was higher than that of 1a,
when 1b was applied to the conversion of 2 to 3, for which the
TON exceeded 105 at room temperature within a few hours.
The high efficiency of 1a,b for the hydrosilane reduction of
amides to enamines or silylhemiaminal intermediates is
satisfactory as a convenient synthesis of enamines in organic
synthesis. However, we suspected that the catalytic perform-
ance of Vaska’s complex 1a could potentially be better than
that actually attained. We observed in experiments with low
catalyst loadings that catalyst deactivation occurred at the later
stage of the reaction. We analyzed the deactivated iridium
species and found that Curtis’ complex 711 was formed.
Interestingly, 7 was reactivated by photolysis. This prompted
us to search for a series of PPh3-based iridium complexes as
catalysts for the conversion of amides to enamines via
silylhemiaminals under thermal or photochemical conditions.
These studies have brought about highly efficient catalysis,
achieving TON >105 in extreme cases, by 1a under photolysis
or by IrCl(PPh3)3 at 60 °C. Consideration of the net
catalytically active species yielded the hypothesis that a “Z-
Ir(PPh3)2” species, where Z is H, Cl, or −SiMe2OSiMe2H, is
possibly involved. This hypothesis provided a convenient
method to access the catalytically active species by simply
mixing [IrCl(η4-COD)]2 with PPh3 or other phosphorus
ligands.
1
single iridium species. H and 31P resonances due to 7 were
observed in the reaction mixture, while the 31P{1H} signal (δ
24.73, s) due to 1a disappeared. An important discovery was
obtained when we photolyzed the reaction mixture by a high-
pressure Hg lamp at room temperature, after the enamine
formation was terminated. As shown in the reaction profile in
Figure 1, after 2 h, the photolysis restarted the catalytic
reaction, though the reaction was much slower than the initial
rate catalyzed by 1a. We also attempted to reactivate the
catalytically active species by heating the reaction mixture after
the catalyst deactivation at 60 °C, which did not restart the
reaction (section 1-1 in the Supporting Information).
We confirmed the catalytic activity of 7, which was isolated
and well-characterized, for the reaction of 2a with TMDS
(Scheme 2). The reaction was slow when the catalyst loading
was 0.001 mol % under photoirradiation conditions. However,
the photochemical treatment of 2a with TMDS in the presence
of 0.05 mol % of 7 in toluene at 25 °C resulted in a 75% yield
of 3a after 1 h. No reaction occurred without photolysis at 25
°C (sections 1-2 and 1-3 in the Supporting Information).
Catalyst Screening for Photochemical and Thermal
Conversion of 2a to 3a. The above results clearly indicate
that formation of 7 is the origin of the catalyst deactivation of
the 1a-catalyzed conversion of amides to enamines with
TMDS, while photolysis of 7 generates a catalytically active
species. Since the photoirradiation of metal carbonyls is known
to result in the dissociation of CO ligands from the metal
center,12 it is likely that regeneration of the catalytically active
species from 7 was promoted by photoassisted dissociation of
CO. These considerations suggested that continuous photo-
irradiation of the reaction of 2a to 3a in the presence of Vaska’s
complex 1a could improve the catalytic efficiency in
comparison with that under dark conditions.
Table 1, entries 2−4, shows the reaction of 2a with TMDS
catalyzed by 0.001 mol % of 1a in toluene. At room
temperature, the reaction was terminated at 25% conversion
of 2a, and no significant improvement of the conversion was
observed at 60 °C (entries 2 and 3). In sharp contrast,
photoirradiation of the reaction solution at room temperature
resulted in efficient production of 3a, for which both the TON
and TOF (h−1) reached 105 (entry 4). Toluene was a better
solvent than dichloromethane and THF (entries 5 and 6).
We considered two possibilities to explain the experimental
results described above: (1) the reaction of 1a with TMDS
during the catalysis is accompanied by formation of inactive 7,
but continuous photolysis can regenerate the catalytic activity
by CO dissociation from 7 (possibility 1); (2) photolysis of
Vaska’s complex 1a results in dissociation of CO to form a
“ClIr(PPh3)2”13 species,12b which may be more catalytically
RESULTS AND DISCUSSION
■
Formation and Reactivation of Curtis’ Complex 7.
The reaction of 1a with TMDS was reported by Curtis and co-
workers in 1977.11 The product was the disilametallacyclic
iridium hydride 7, as shown in Scheme 2. Two moles of
TMDS was involved in the formation of 7; one molecule of
TMDS contributed to the exchange of a chloride ligand in 1a
by a hydride, and double oxidative addition of another
Scheme 2. Catalytic Activity of Curtis’s Complex 7
B
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 1. Hydrosilane Reduction of PhCH2CONEt2 (2a) To Form PhCHCHNEt2 (3a) Catalyzed by 1a and 8 under Thermal
a
and Photochemical Conditions
b
b
entry
catalyst
mol %
Δ or hν
temp (°C)
solvent
conversn of 2a (%)
yield of 3a (%)
1
2
3
4
5
6
7
8
IrCl(CO)(PPh3)2 (1a)
1a
1a
1a
1a
1a
0.05
Δ
Δ
Δ
hν
hν
hν
Δ
Δ
Δ
25
25
60
25
25
25
25
25
60
60
60
25
toluene
toluene
toluene
toluene
CH2Cl2
THF
toluene
toluene
toluene
CH2Cl2
THF
99
25
30
99
82
48
99
14
81
41
20
34
99
25
30
99
81
48
99
14
81
41
20
34
0.001
0.001
0.001
0.001
0.001
0.05
0.001
0.001
0.001
0.001
0.001
IrCl(PPh3)3 (8)
8
8
8
8
8
9
10
11
12
Δ
Δ
hν
toluene
a
All reactions were carried out by using amide 2a (1.00 mmol), TMDS (2.0 mmol), and the solvent given above (0.5 mL total) in the presence of
b
1a or 8 (0.05 mol % or 0.001 mol %) under an inert gas atmosphere for 1 h. Determined by 1H NMR in the presence of anisole as the internal
standard.
catalysts to convert 2a to 3a. The list of the Ir-PPh3 catalysts
contains IrH(CO)(PPh3)3 (9),15 IrH(η4-COD)(PPh3)2
(10),16 IrH5(PPh3)2 (11),17 IrH3(PPh3)3 (12),18 and a
disilyltrihydride iridium(V) compound (13),19 which has a
disilametallacyclic skeleton derived from BDSB. Corriu and co-
workers briefly reported the reaction of 2a with BDSB to form
3a by catalysis of RhCl(PPh3)3 (14), in which the catalytic
activity of 14 was much lower than that of 8.20 We examined
the catalytic activity of RhCl(CO)(PPh3)2 (15),21 RhH(CO)-
(PPh3)3 (16),22 and RhH(PPh3)4 (17)22 in addition to 14.
The results are summarized in Chart 1, and the details are
described in sections 1-2 through 1-5 in the Supporting
Information. 9−17 all showed some catalytic activity, but none
of them exhibited higher activity in comparison to photo-
chemically activated 1a and thermally activated 8. However,
four of them promoted the reaction with 0.5−0.05 mol % of
Figure 1. Reaction profile for the reaction of PhCH2CONEt2 (2a)
with TMDS in the presence of 0.001 mol % of IrCl(CO)(PPh3)2
(1a).
active than 1a itself or the species photochemically generated
from 7 (possibility 2). As described above, the catalytically
active species generated by photolysis of 7 was much less
efficient than that from 1a in the dark. Since the catalytic
activity of 1a under photolysis in toluene (entry 4) was higher
than we expected from this, we assumed possibility 2 to be
more reasonable.
Chart 1. Catalytic Activity of a Series of Ir/Rh Catalysts for
Hydrosilane Reduction of 2a with TMDS
We were interested in an alternative generation method of
“ClIr(PPh3)2”, which is formed by photoirradiation of
IrCl(CO)(PPh3)2, and thought that thermal dissociation of
PPh3 from IrCl(PPh3)3 (8)14 could be useful. As shown in
Table 1, entry 7, the catalytic activity of 8 was as high as that of
1a at room temperature, when 0.05 mol % of 8 was used as the
catalyst. While the reaction with 0.001 mol % of 8 resulted in
low conversion of 2a to 3a (14%) at room temperature for 1 h
(entry 8), heating at 60 °C increased the conversion of 2a
(entry 9) 6 times as much as that at 25 °C; the catalyst
efficiency is high enough to be comparable to that of 1a/hν
shown in entry 4. Among three solvents examined, toluene was
better than CH2Cl2 and THF (entries 10 and 11). It was
confirmed that photoirradiation of the reaction mixture at
room temperature did not contribute to increase the efficiency
(entry 12). In the cases of both 1a and 8, TMDS was the best
hydrosilane in comparison with PMDS, BDSB, and PhMe2SiH
(sections 1-6 and 1-7 in the Supporting Information).
This discovery of the highly active Ir-PPh3-based catalyst
systems, photochemically activated 1a and thermally activated
8, prompted us to screen additional iridium complexes for
C
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Hydrosilane Reduction of N,N-Dialkyl- or N-Alkyl-N-arylamides Catalyzed by Photochemically Activated 1a or
a
Thermally Activated 8 with 0.001 mol % Catalyst Loading
a
All reactions were carried out by using amides 2 (1.00 mmol), TMDS (2.0 mmol), toluene (0.5 mL total) in the presence of 1a or 8 (0.001 mol
b
1
%) under inert gas atmosphere. Determined by H NMR in the presence of anisole as the internal standard.
Table 3. Hydrosilane Reduction of N,N-Diarylamides Catalyzed by Photochemically Activated 1a or Thermally Activated 8
a
with 0.5 mol % Catalyst Loading
a
All reactions were carried out by using amides 4 (1.00 mmol), TMDS (2.0 mmol), toluene (0.5 mL total) in the presence of 1a or 8 (0.5 mol %)
b
1
under inert gas atmosphere. Determined by H NMR in the presence of anisole as the internal standard.
D
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 4. Hydrosilane Reduction of Various Amides 2 and 4 To Form Enamines 3 and 6 by in Situ Generation of the
a
Catalytically Active Species from [IrCl(η4-COD)Cl]2 and PPh3 or P(OC6F5)3
product [NMR yield
b
(%)]
b
entry
amide
cat. (mol % per Ir atom)
L
time (h)
conversn of 5 (%)
5
3 or 6
isolated yield of 6 (%)
c
1
2
3
4
5
6
2a
2d
2e
4a
4c
0.001
0.001
0.001
0.5
0.5
0.5
PPh3
PPh3
PPh3
PPh3
PPh3
P(OC6F5)3
2
2
6
4
24
99
99
99
99
0
3a [99]
3d [99]
3e [99]
6a [99]
6c [0]
96 (99)
5a [0]
5c [0]
5c [99]
92
0
84 (86)
d
2
99
6c [0]
a
All reactions were carried out by using amides 2 (1.00 mmol), TMDS (2.0 mmol), and toluene (0.5 mL total) in the presence of [IrCl(η4-
b
COD)Cl]2 (0.0005 mol % or 0.25 mol %) and PPh3 (0.002 mol % or 1.0 mol %) under an inert gas atmosphere. Determined by 1H NMR in the
presence of anisole as the internal standard. Isolated yield when the reaction was performed with 0.1 mol of 2a. Isolated yield when the reaction
was performed with 10 mmol of 4c.
c
d
catalyst loading both thermally and photochemically. Lower
catalytic activity was observed for the iridium complex 12 and
all of the rhodium complexes. Although the catalyst screening
did not yield any catalysts more efficient than 1a/hv or 8a at 60
°C, the results are helpful for the mechanistic considerations
described later.
However, the attempted isolation of 3e was not successful,
giving a complicated mixture of products.
As reported previously, N,N-diarylamides 4 are precursors of
π-conjugated enamines 6. Electrons on the amide groups are
delocalized over the aryl groups on the nitrogen atom, which
makes the reaction with TMDS by catalysis of IrCl(CO)-
(PPh3)2 difficult. The reactivity of 2a and four N,N-
diarylamides decreased in the order 2a ≫ 4a > 4b > 4c, 4d.9b
Under the conditions generally used for the reactions of the
usual amides 2 (0.5 mol % of 1a, at room temperature for 2 h),
the reactions of 4b−d did not occur, but they did take place by
using 1b as the catalyst (0.5 mol %, at room temperature for 2
h).9b The primary product of the reaction of 4 was not the
enamine 6 but the silylhemiaminal 5, which was converted to 6
under acidic conditions or by heat. Table 3 shows the reactions
of the four amides 4a−d with TMDS catalyzed by photo-
activated 1a at room temperature or 8 at 60 °C. With 0.5 mol
% catalyst loading, the reactions of 4a were over within 2 h
using both systems, but the products were different. The
silylhemiaminal 5a was formed as a single product with the 1a/
hν system, which was then converted to enamine 6a by
thermal treatment at 100 °C. In contrast, when 8 was used as
the catalyst at 60 °C, 5a was not observed at the end of the
reaction but 6a was formed selectively. The reaction of amide
4b with TMDS by catalysis of the 1a/hν system also gave the
corresponding silylhemiaminal 5b selectively, whereas the
reaction with 8 at 60 °C gave a mixture of 5b and 6b. In both
cases, conversion of 4b reached over 97% after several hours,
and thermal treatment of the obtained crude product gave the
enamine 6b in good yields. The reactions of 4c,d with TMDS
also occurred with the 1a/hν system at 25 °C and 8 at 60 °C.
This is in sharp contrast to the previous results in which no
reaction with IrCl(CO)(PPh3)2 occurred at 25−80 °C.
However, the reactions were terminated after 25−76% of the
starting material was consumed, and attempts to complete the
reaction have so far been unsuccessful. Use of IrCl(CO)[P-
(OC6F5)3]2 (1b) as the catalyst for the reactions of 4c,d was
more efficient than those using the 1a/hν system at 25 °C or 8
at 60 °C as the catalyst.9b
Scope of the Substrates for the Reactions Catalyzed
by 1a/hν or 8 at 60 °C. The catalysis of photoactivated 1a
and thermally activated 8 provided excellent results in the
conversion of a variety of tertiary amides to the corresponding
enamines. Table 2 shows the results for the reactions of five
N,N-dialkyl- and N-alkyl-N-arylamides 2a−e with TMDS. All
of the reactions were carried out with 0.001 mol % catalyst
loading. Both the conversion of the amide and the yield of the
product were determined by 1H NMR spectroscopy, and
isolated yields are reported in the table. With either 1a at 25
°C under photolysis or 8 at 60 °C, two phenylacetamide
derivatives, 2a,b, underwent the hydrosilane reduction and
subsequent elimination of silanol to form 3a,b, respectively, in
quantitative yields. As described previously, a phenyl group on
the amide nitrogen led to lowered activity of amides toward
the iridium-catalyzed hydrosilane reduction.9b For this reason,
the reaction with 2b needed a somewhat longer reaction time
to complete the reaction than did 2a. The aliphatic amide 3c
was also smoothly converted to the corresponding enamine, as
shown in entries 5 and 6.
A characteristic feature of these iridium-catalyzed hydro-
silane reductions of amides to enamines is the high functional
group compatibility. Table 2, entries 7−10, show the reactions
of amides including another reducible functional group such as
a carbonyl group in the molecule. Amide ester 2d was
converted to the corresponding enamino ester 3d catalyzed by
either 1a under photolysis or 8 at 60 °C. After 6 h, the crude
product was unstable, but it could be isolated by passage
through Celite followed by washing with cold pentane. The
experiment shown in entry 8 afforded 3d in 86% yield. Keto
amide 2e was also subjected to the catalysis of photochemically
activated 1a or thermally activated 8 for 4 h. In both cases,
NMR spectra of the crude sample showed quantitative
formation of 3e, as shown in the Supporting Information.
E
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
a
In Situ Generation of Catalytically Active Species. It is
known that IrCl(PPh)3 (8) was prepared by treatment of
[IrCl(η4-COD)]2 with PPh3.14 As described above, enamines
were formed from amides by using 8 as the catalyst, and the
active species “ClIr(PPh3)2” was formed by thermal dissoci-
ation of PPh3 from 8. We believed that these catalytically active
species could be conveniently generated in situ from a mixture
of [IrCl(η4-COD)]2 with PPh3. One advantage of this in situ
generation method is the possible use of various phosphorus
ligands instead of PPh3. Table 4 clearly demonstrates that our
idea provided fruitful outcomes. The results shown in entries
1−4 using a mixture of [IrCl(η4-COD)]2 with PPh3 (Ir:P =
1:2) correspond to the experiments reported as entries 2, 8,
and 10 in Table 2 and entry 2 in Table 3 using IrCl(PPh3)3
(8). It is clear that a mixture of [IrCl(η4-COD)]2 with PPh3
(Ir:P = 1:2) behaved as a good catalyst comparable to 8. As
shown in entry 6 in Table 3, the reaction of the amide 4c was
not complete with 8 at 60 °C for 24 h. The in situ generated
catalyst from [IrCl(η4-COD)]2 with PPh3 was even less active
than 8, affording no conversion of 4c at 60 °C for 24 h. In
contrast, full conversion of 4c and quantitative formation of 5c
were achieved with a mixture of [IrCl(η4-COD)]2 with
P(OC6F5)3 (Ir:P = 1:2).
Scheme 3. Possible Catalytic Cycles
The synthetic utility of this in situ generation method was
demonstrated by the large-scale production of two enamines.
The first is an example of N,N-dialkylenamine synthesis on a
0.1 mol scale; the amide 2a (19.1 g) and TMDS (26.8 g) were
subjected to catalysis with a mixture of [IrCl(η4-COD)]2 and
PPh3 (Ir:P = 1:2, 0.001 mol % per Ir atom) at 60 °C for 19 h
to give 3a (17.4 g, 99.2 mmol) in 99% isolated yield. The
second is the preparation of a N,N-diarylenamine on a 1 g
scale; treatment of 4c (1.43 g) with TMDS (1.34 g) in the
presence of a mixture of [IrCl(η4-COD)]2 and P(OC6F5)3
(Ir:P = 1:2, 0.5 mol % per Ir atom) at 60 °C for 2 h gave 3a
(1.16 g) in 86% isolated yield.
Considerations of the Active Species Generated from
“ClIr(PPh3)2” and TMDS. As described above, the catalytic
conversion of amides to enamines occurs stepwise, and the
iridium catalysis only takes part in the first step: i.e., the
hydrosilylation of amides to silylhemiaminals. The results
described in this paper are in good agreement with the
hypothesis that the 14-electron species “ClIr(PPh3)2”, which is
stabilized by solvents or reactants in the actual catalytic
reactions,13 is generated from the catalyst precursors,
contributing to the catalytic hydrosilylation reaction. In fact,
it is known that “ClIr(PPh3)2” is generated by photolysis of
IrCl(CO)(PPh3)2 (1a).12b From analogy to RhCl(PPh3)3,
“ClIr(PPh3)2” can be generated by thermal dissociation of
PPh3 from IrCl(PPh3)3 (8).23 The results whereby a mixture of
[IrCl(η4-COD)]2 with PPh3 (Ir:P = 1:2) behaved as a good
catalyst comparable to 8 support the involvement of
“ClIr(PPh3)2” at any step in the catalytic reaction (Scheme 3a).
In contrast, the following discussion indicates that the Cl−Ir
bond is not necessarily important for the catalytically active
species. It is known that a metal−Cl bond of a transition-metal
complex is converted to a metal hydride and a metal silyl
species by the action of a hydrosilane.24 As shown in Scheme
3a, treatment of “ClIr(PPh3)2” with a Si−H moiety affords
“HIr(PPh3)2”.13 The reaction of “HIr(PPh3)2” with a Si−H
bond is reversible with that of “SiIr(PPh3)2”13 with H2 by way
of a “SiIrH2(PPh3)2” intermediate. Supporting evidence for
Scheme 3a is seen in the formation mechanism of Curtis’
complex 7. The Cl/H exchange is involved in the reaction of
a
Legend: (*) deactivation process when 1a is used as the catalyst;
(**) reactivation process from 7 under photoirradiation.
1a with TMDS to form “HIr(PPh3)2” and ClMe2SiOSiMe2H,
which proceeds through the oxidative addition of a Si−H bond
in TMDS to “ClIr(PPh3)2” and subsequent reductive
elimination of a Si−Cl bond.11 The H/Si exchange is seen in
further reaction of the resulting “HIr(PPh3)2” species with
TMDS. This forms a “SiIrH2(PPh3)2” intermediate, from
which reductive elimination of H2 forms the “SiIr(PPh3)2”
species, where Si = Me2HSiOMe2HSi−. The formation of 7 is
accomplished by intramolecular oxidative addition of the Si−H
moiety in “Me2HSiOMe2Si−Ir(PPh3)2”. In other words, in
addition to “ClIr(PPh3)2”, “HIr(PPh3)2” and “SiIr(PPh3)2”
species could be involved in the reaction mechanism. The
involvement of iridium hydride or iridium silyl species as the
catalytically active species is supported by the experimental
results obtained from the catalyst screening. As described in
the sections 1-2 through 1-5 in the Supporting Information in
detail, we studied several iridium hydride and silyl iridium
complexes, IrH5(PPh3)2 (10), IrH3(PPh3)3 (12), IrH(η4-
COD)(PPh3)2 (9), and one disilyl trihydride iridium(V)
compound, 13, having a disilametallacyclic skeleton derived
from BDSB as candidates for alternative efficient catalysts. All
of these compounds more or less showed catalytic activity
toward the reaction of 2a with TMDS to form 3a as a single
product, though none of them were as efficient as 1a under
F
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
photolysis or 8 at 60 °C. Two chloro complexes, RhCl(PPh3)3
(14) and RhCl(CO)(PPh3)2 (15), showed some activity, but
the rhodium hydride catalysts RhH(CO)(PPh3)3 (16) and
RhH(PPh3)4 (17) also behaved as catalysts. It is known that
the Rh−Cl bond is converted to a Rh−H bond by the action of
a hydrosilane.24 From these results, we consider that
“ClIr(PPh3)2” is generated from 1a, 8, and a mixture of
[IrCl(η4-COD)]2 with PPh3 (Ir:P = 1:2) and exists in the
reaction medium. However, it does not participate directly in
the catalytic cycle but rather acts as an intermediary species to
generate “HIr(PPh3)2” and “SiIr(PPh3)2” species.
in this species forms the 18e Ir(V) disilametallacyclic species
X2. If X1 is the active species, X2 is a dormant species
reversibly generating X1. X2 is also the source of the
deactivated species 7, which can be formed by reductive
elimination of H2 from X2 followed by coordination of CO. As
described above, regeneration of the active species from 7 is
achieved by photoirradiation. These deactivation/reactivation
steps would be involved in the reactions using 1a/hv.32 It is
worthwhile to point out that the Ir(V) species X3 is possibly
generated from X2 by dissociation of PPh3, which may be
another candidate for the catalytic intermediate.
Regarding the selectivity of the product, this is in sharp
contrast to the results that ruthenium-, platinum-, and iron-
catalyzed hydrosilane reductions of amides result in selective
formation of amines.5−7 Nakatani’s calculation suggested the
involvement of H2PtIVSi2, of which one of the two hydrides
and one of the two silyl species take part in rapid
hydrosilylation of an amide to a silylhemiaminal. The resulting
HPtIISi species contributes to conversion of the silylhemiami-
nal to a mixture of a disiloxane and an amine.31 In the case of
the present Ir-catalyzed reactions of amides with TMDS, the
formation of silylhemiaminals is rapid but further conversion of
the silylhemiaminals to the amines and the disiloxanes is
somehow suppressed.
Scheme 3b illustrates a simplified catalytic cycle for the
hydrosilylation of amides. Mechanisms of hydrosilylation of
alkenes were investigated by Sakaki and co-workers, who
proposed the involvement of oxidative addition of a Si−H
bond to either Pt(PH3)2 or RhCl(PH3)2 at the initial step of
the catalytic cycle.25 26c,d In a similar fashion, oxidative
c,
addition of a Si−H bond to Z-Ir(PPh3)2, where Z = Cl, H,
Si, would initiate the iridium-catalyzed hydrosilylation of
amides, which forms an Ir(III) silyl hydride species, Z-
Ir(H)Si(PPh3)2. Insertion of a CO bond in the amide
between the Ir−H bond or between the Ir−Si bond initiates
the Chalk−Harrod25 and modified Chalk−Harrod26 cycles,
respectively. The resulting insertion products undergo the
reductive elimination of a silylhemiaminal to regenerate Z-
Ir(PPh3)2.27−29
Chalk−Harrod and modified Chalk−Harrod cycles medi-
ated by Z-Ir(PPh3)2 are general schemes for hydrosilylation;
however, they cannot explain two characteristic features in the
catalytic hydrosilane reduction catalyzed by iridium phosphine
or iridium phosphite catalysts. The first is the extremely high
catalyst efficiency, which is achieved only with TMDS. The
second is the selective formation of enamines from amides by
way of silylhemiaminal intermediates.
Regarding the high catalytic activity, we have published a
series of papers in which unusual rate enhancement by two
closely located Si−H groups in bifunctional hydrosilanes was
generally observed in hydrosilylation of ketones and aldehydes
to the corresponding silyl ethers or the hydrosilane reduction
of amides to amines.4,30 We carried out mechanistic studies on
this proximity effect of the two Si−H groups in the platinum-
catalyzed hydrosilane reduction of amides with BDSB and
isolated several disilaplatina(IV)cycles, which were formed by
double oxidative addition of two Si−H groups in BDSB to the
Pt(0) precursor and showed catalytic activity toward hydro-
silane reduction of amides to amines.6 Nakatani and co-
workers performed DFT calculations based on these
experimental results and proposed that the H2PtIVSi2 species
is responsible for the high catalytic activity and selective
formation of amines.31 Experimental studies to isolate dormant
species were performed for the RhCl(PPh3)3-catalyzed hydro-
silylation of acetone, which suggest that double oxidative
addition of two Si−H groups in BDSB to the rhodium center is
involved in the catalytic hydrosilylation of acetone to the
corresponding isopropoxysilane. A H3RhSi2 species was
identified by NMR spectroscopy, suggesting that the Rh−Cl
bond in the rhodium species is replaced by a hydride.24
In this context, the scheme shown in Scheme 3c should be
considered as the key to understanding the high reactivity of
the iridium-catalyzed hydrosilylation of amides to silylhemia-
minals. Oxidative addition of one of the Si−H bonds in TMDS
gives a 16e species, (Me2HSiOMe2Si)Ir(H)(Z)L2 (X1; Ir(III),
L = PPh3). Intramolecular oxidative addition of the Si−H bond
Detailed studies on the catalytically active species, possible
reaction pathways, and explanation of unusually high
hidrosilylation activity and selective formation of silylhemia-
mals from amides require further investigation and are now
underway by using DFT calculations in collaboration with Dr.
Nakatani.33
CONCLUSION
■
Two new methods for the efficient conversion of tertiary
amides to enamines were established by the discovery of
photochemically activated IrCl(CO)(PPh3)2 (1a) and ther-
mally activated IrCl(PPh3)3 (8) as species with extremely high
catalytic activity. In extreme cases, the TON and TOF (h−1)
reached 105 with full conversion of the starting amide.34 These
new catalysts contribute to the successful synthesis of a variety
of enamines in high yields with low catalyst loadings. Unique
functional group compatibility leads to the selective reaction of
amides with coexisting ester or ketone groups remaining intact.
Mechanistic studies indicate that the 14-electron species
“ClIr(PPh3)2” is generated from 1a/hν or 8, which is a
precursor of “HIr(PPh3)2” and “SiIrCl(PPh3)2” species
involved in the catalytic cycle. An alternative source of
“ClIr(PPh3)2” is a mixture of commercially available [IrCl(η4-
COD)]2 and PPh3, which generate in situ a highly catalytically
active species useful for conversion of N,N-dialkylamides to
enamines. A mixture of [IrCl(η4-COD)]2 and P(OC6F5)3
generates “ClIr[P(OC6F5)3]2”, which contributes to the facile
synthesis of π-conjugated enamines from N,N-diarylamides.
The synthetic utility was demonstrated by large-scale (10
mmol ∼0.1 mol) syntheses of two enamines.
The present method is useful for practical syntheses of
enamines from easily available iridium complexes. The new
catalysts including in situ generation of active species showed
distinguished catalytic activity and interesting functional group
compatibility. The IrCl(CO)(PPh3)2-catalyzed reactions of
tertiary amides with TMDS are now utilized by Chida, Dixon,
and other chemists as in situ generation of silylhemiaminals,
which reacted with nucleophiles to synthesize functionalized
amino derivatives including natural products. The present
G
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
results contribute to the progress of this research field.35 In
addition to the synthetic utility, mechanistic considerations
provide fresh insights that include the involvement of
“HIr(PPh3)2” and “SiIrCl(PPh3)2” species in the catalytic
cycle. Elucidation of detailed mechanisms by DFT calculations
is now in progress.
mg, 1.0 mmol, as an internal standard), and TMDS (268.7 mg, 2.0
mmol) were added. The mixture was irradiated with a high-pressure
mercury lamp (entries 1, 3, 5, 7, and 9) or heated at 60 °C (entries 2,
4, 6, 8, and 10) for the time described in the table. The conversion of
1
2 and the yield of 3 were determined by H NMR spectroscopy. For
2a,b, the reaction mixture was filtered through a pad of Celite and the
resulting silane residue including unreacted TMDS and the solvent
were removed in vacuo to obtain enamines 3a,c as colorless oils. For
2d, after the same treatment as that of 2a−c, the resulting sample was
cooled and washed with cold hexane (−30 °C) followed by
evacuation to obtain 3d. For 2c,e, the corresponding enamines 3c,e
were too unstable due to the strong density to be isolated; therefore,
only the NMR yield was determined.
Hydrosilane Reduction of N,N-Diarylamides 4 To Form the
Corresponding π-Conjugated Aldenamines 6 Catalyzed by
Photochemically Activated 1a or Thermally Activated 8
(Table 3). In a two-necked 20 mL round flask were placed complex
1a (3.9 mg, 5.0 × 10−3 mmol) or 8 (4.5 mg, 5.0 × 10−3 mmol),
toluene (0.5 mL), a substrate 4a−4d (1.0 mmol), anisole (108.1 mg,
1.0 mmol, as an internal standard), and TMDS (268.7 mg, 2.0 mmol).
The mixture was irradiated with a high-pressure mercury lamp
(entries 1, 3, 5 ,and 7) or heated at 60 °C (entries 2, 4, 6, and 8) for
the time described in the table. The conversion of 4 and the yield of 5
or 6 were determined by 1H NMR spectroscopy. For 4a,b, the
reaction mixture was filtered through a pad of Celite and the resulting
silane residue including unreacted TMDS and the solvent were
removed in vacuo. The obtained viscous solid was again dissolved in
toluene (10 mL) and heated at 80 °C for 2 h to convert 5 to 6
according to a previous report.9b The solution was concentrated
under reduced pressure. Purification by silica gel column chromatog-
raphy (hexane/toluene) gave 6a,b.
EXPERIMENTAL SECTION
■
General Procedure. All reactions were carried out under an
argon atmosphere using standard Schlenk techniques or performed in
1
a N2-filled glovebox. H, 13C, 19F, 29Si, and 31P NMR spectra were
measured on JEOL ECA 400 (396 MHz) and ECA600 (600 MHz)
1
spectrometers. Chemical shifts for H and 13C are given in parts per
million relative to the solvent signal. Chemical shifts for 31P are given
in parts per million downfield from phosphoric acid (δP 0) as an
external standard. IR spectra were measured on a JASCO FT/IR 4200
spectrometer. HR-MS analyses were performed at the Analytical
Center at the Institute for Materials Chemistry and Engineering,
Kyushu University. Photoirradiation was carried out by an Ushio UM-
453B-A high-pressure mercury lamp (450 W) according to the
procedure below; a reaction mixture was placed in a VIDTEC 30 mL
quartz round-shaped flask equipped with a three-way stopcock. The
flask was set at a position of 15 cm away from the lamp. A quartz
dewar condenser was also set around the lamp to avoid thermal
effects. Then the solution was stirred under photoirradiation for the
time described in Tables 1−3. For the reaction on an NMR scale (see
sections 1-1, 1-3, and 1-5 in the Supporting Information), a Pyrex
NMR tube equipped with a J. Young valve was used as a reaction
vessel. Deuterated solvents (C6D6, CDCl3) were purchased from
Wako Pure Chemical Industries Ltd. and used after dehydration by
using Na/Ph2CO (for C6D6) or activated MS4A (for CDCl3).
Hydrosilane reagents, 1,1,3,3-tetramethyldisiloxane (TMDS),
1,1,3,3,3-pentamethyldisiloxane (PMDS), and PhMe2SiH were
purchased from Wako Pure Chemical Industries, AZmax Co. Ltd.,
respectively, and distilled before using. Dehydrated solvents (n-
pentane, toluene, THF, ethyl acetate, CH2Cl2), PPh3, and anisole
were purchased from Kanto Chemical Co. Ltd. and used as received.
Iridium complexes 1a,36 7,11 8,14 9,15 10,16 11,17 12,18 13,19 and
[IrCl(η4-COD)]2,37 rhodium derivatives 14,20 15,21 16,22 and 17,22
the phosphite reagent P(OC6F5)3,38 amides 2a−e,9a 4a,39 and 4b−
d,9b and the hydrosilane reagent 1,2-bis(dimethylsilyl)benzene24 were
prepared according to the literature.
Hydrosilane Reduction of 4 To Form 6 by in Situ
Generation of the Catalytically Active Species from [IrCl(η4-
COD)]2 and PPh3 or P(OC6F5)3 (Table 4). A stock solution of the
catalyst was prepared by dissolving [IrCl(η4-COD)]2 (3.4 mg, 5.0 ×
10−3 mmol) and PPh3 (5.2 mg, 2.0 × 10−2 mmol) or P(OC6F5)3
(12.8 mg, 2.0 × 10−2 mmol) in toluene (10 mL) and stirring for 10
min, followed by the dilution of 0.1 mL of the solution with 0.9 mL of
toluene. A 0.1 mL portion of the stock solution (0.001 mol % per Ir
atom) was placed in a two-necked 20 mL round flask, and then
toluene (0.4 mL), one of the substrates 2d,e and 4a,c (1.0 mmol),
anisole (108.1 mg, 1.0 mmol, as an internal standard), and TMDS
(268.7 mg, 2.0 mmol) were added. The mixture was stirred at 60 °C
for the time described in the table. The conversion of amides and the
1
yield of enamines or silylhemiaminals were determined by H NMR
Hydrosilane Reduction of PhCH2CONEt2 (2a) To Form
PhCHCHNEt2 (3a) Catalyzed by 1a and 8 under Thermal
and Photochemical Conditions (Table 1). For entries 1 and 7, a
stock solution of the catalyst was prepared by dissolving complex 1a
(7.8 mg, 1.0 × 10−2 mmol) or 8 (10.1 mg, 1.0 × 10−2 mmol) in
toluene (2 mL). For entries 2−6 and 8−12, a stock solution of the
catalyst was prepared by dissolving complex 1a (7.8 mg, 1.0 × 10−2
mmol) or 8 (10.1 mg, 1.0 × 10−2 mmol) in toluene, THF, or CH2Cl2
(10 mL), followed by the dilution of 0.1 mL of the solution with 0.9
mL of the solvent. A 0.1 mL portion of the stock solution of 1a or 8
(0.05 mol % for entries 1 and 7, 0.001 mol % for entries 2−6 and 8−
12) was placed in a two-necked 20 mL round flask, and then solvent
(0.4 mL), substrate 2a (191.2 mg, 1.0 mmol), anisole (108.1 mg, 1.0
mmol, as an internal standard), and TMDS (268.7 mg, 2.0 mmol)
were added. The mixture was stirred at 25 °C (entries 1, 2, 7, and 8),
or 60 °C (entries 3 and 9−11) or irradiated with a high-pressure
mercury lamp (entries 4−6 and 12) for 1 h. The conversion of 2a and
spectroscopy. For 4a, the reaction mixture was filtered through a pad
of Celite and the resulting silane residue including unreacted TMDS
and the solvent were removed in vacuo. Purification by silica gel
column chromatography (hexane/toluene) gave 6a. For 4c, after the
same treatment as that of 4a, toluene (10 mL) and P2O5 (30 mg)
were added to the resulting mixture and stirred at 25 °C for 0.5 h to
convert 5c to 6c according to a previous report.9b The solution was
concentrated under reduced pressure. Purification by silica gel column
chromatography (hexane/toluene) gave 6c.
The 0.1 mol Scale Reaction of 2a with TMDS Catalyzed by
in Situ Generated Catalyst Species from [IrCl(η4-COD)]2 and
PPh3 (Entry 1 in Parentheses, Table 4). A stock solution of the
catalyst was prepared by dissolving [IrCl(η4-COD)]2 (3.4 mg, 0.5 ×
10−2 mmol) and PPh3 (5.2 mg, 2.0 × 10−2 mmol) in toluene (10 mL)
After the mixture was stirred for 10 min, 1.0 mL of the stock solution
(0.001 mol % per Ir atom) was placed in a two-necked 100 mL round
flask that contained amide 2a (19.1 g, 100.0 mmol) and toluene (9
mL). TMDS (26.8 g, 200.0 mmol) was then placed slowly into the
flask. The mixture was stirred at 60 °C for 19 h. Complete
1
the yield of 3a were determined by H NMR spectroscopy.
Hydrosilane Reduction of N,N-Dialkylamides 2 To Form the
Corresponding Aldenamines 3 Catalyzed by Photochemically
Activated 1a or Thermally Activated 8 (Table 2). A stock
solution of the catalyst was prepared by dissolving complex 1a (7.8
mg, 1.0 × 10−2 mmol) or 8 (10.1 mg, 1.0 × 10−2 mmol) in toluene
(10 mL), followed by the dilution of 0.1 mL of the solution with 0.9
mL of the solvent. A 0.1 mL portion of the stock solution of 1a or 8
(0.001 mol %) was placed in a two-necked 20 mL round flask, and
then toluene (0.4 mL), the substrate 2a−e (1.0 mmol), anisole (108.1
1
consumption of the substrate was confirmed by H NMR spectros-
copy. After removal of the volatiles, the product 3a was obtained by
distillation under reduced pressure (8 Pa, 100 °C) as a colorless oil
(17.4 g, 99.2 mmol, 99%).
Reaction of 4c with TMDS (10 mmol Scale) Catalyzed by in
Situ Generated Catalyst Species from [IrCl(η4-COD)]2 and
P(OC6F5)3 (Entry 6 in Parentheses, Table 4). In a two-necked 50
H
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
mL round flask were placed [IrCl(η4-COD)]2 (8.5 mg, 1.25 × 10−2
mmol), P(OC6F5)3 (29.0 mg, 5.00 × 10−2 mmol), toluene (2.5 mL),
and amide 4c (1.43 g, 5.0 mmol). TMDS (1.34 g, 10.0 mmol) was
then placed slowly into the flask. The mixture was heated at 60 °C for
2 h. The reaction mixture was filtered through a pad of Celite, and the
resulting silane residue including unreacted TMDS and the solvent
were removed in vacuo. The resulting mixture was dissolved in
toluene (50 mL), and P2O5 (150 mg) was added to the solution at
−78 °C and stirred at 25 °C for 1 h to convert 5c to 6c. The solution
was concentrated under reduced pressure.9b Purification by silica gel
column chromatography (hexane/toluene) gave 6c (1.16 g, 4.3 mmol,
86%).
Triruthenium Carbonyl Cluster Bearing a Bridging Acenaphthylene
Ligand: An Efficient Catalyst for Reduction of Esters, Carboxylic
Acids, and Amides by Trialkylsilanes. J. Org. Chem. 2002, 67, 4985−
4988.
(4) (a) Nagashima, H. Efficient Transition Metal-Catalyzed
Reactions of Carboxylic Acid Derivatives with Hydrosilanes and
Hydrosiloxanes, Afforded by Catalyst Design and the Proximity Effect
of Two Si-H Groups. Synlett 2015, 26, 866−890. (b) Sunada, Y.;
Nagashima, H. Disilametallacyclic chemistry for efficient catalysis.
Dalton Trans. 2017, 46, 7644−7655.
(5) (a) Hanada, S.; Ishida, T.; Motoyama, Y.; Nagashima, H. The
ruthenium-catalyzed reduction and reductive N-alkylation of secon-
dary amides with hydrosilanes: practical synthesis of secondary and
tertiary amines by judicious choice of hydrosilanes. J. Org. Chem.
2007, 72, 7551−7559. (b) Motoyama, Y.; Mitsui, K.; Ishida, T.;
Nagashima, H. Self-Encapsulation of Homogeneous Catalyst Species
into Polymer Gel Leading to a Facile and Efficient Separation System
of Amine Products in the Ru-Catalyzed Reduction of Carboxamides
with Polymethylhydrosiloxane (PMHS). J. Am. Chem. Soc. 2005, 127,
13150−13151. (c) Reeves, J. T.; Tan, Z.; Marsini, M. A.; Han, Z. S.;
Xu, Y.; Reeves, D. C.; Lee, H.; Lu, B. Z.; Senanayake, C. H. A
Practical Procedure for Reduction of Primary, Secondary and Tertiary
Amides to Amines. Adv. Synth. Catal. 2013, 355, 47−52.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.8b00835.
■
S
Experimental details, spectral data for complexes and
compounds, and NMR spectra (PDF)
AUTHOR INFORMATION
Corresponding Author
*H.N.: e-mail, nagasima@cm.kyushu-u.ac.jp; tel, +81-92-583-
7819; fax, +81-92-583-7819.
■
(6) (a) Hanada, S.; Motoyama, Y.; Nagashima, H. Dual Si-H effects
in platinum-catalyzed silane reduction of carboxamides leading to a
practical synthetic process of tertiary amines involving self-
encapsulation of the catalyst species into the insoluble silicone resin
formed. Tetrahedron Lett. 2006, 47, 6173−6177. (b) Hanada, S.;
Tsutsumi, E.; Motoyama, Y.; Nagashima, H. Practical access to
amines by platinum-catalyzed reduction of carboxamides with
hydrosilanes: Synergy of dual Si-H groups leads to high efficiency
and selectivity. J. Am. Chem. Soc. 2009, 131, 15032−15040.
(7) (a) Sunada, Y.; Kawakami, H.; Imaoka, T.; Motoyama, Y.;
Nagashima, H. Hydrosilane Reduction of Tertiary Carboxamides by
Iron Carbonyl Catalysts. Angew. Chem., Int. Ed. 2009, 48, 9511−9514.
(b) Sunada, Y.; Tsutsumi, H.; Shigeta, K.; Yoshida, R.; Hashimoto,
T.; Nagashima, H. Catalyst design for iron-promoted reductions: an
iron disilyl-dicarbonyl complex bearing weakly coordinating η2-(H-Si)
moieties. Dalton Trans. 2013, 42, 16687−16692. (c) Tsutsumi, H.;
Sunada, Y.; Nagashima, H. New catalyst systems for iron-catalyzed
hydrosilane reduction of carboxamides. Chem. Commun. 2011, 47,
6581−6583. (d) Zhou, S.; Junge, K.; Addis, D.; Das, S.; Beller, M. A
convenient and general iron-catalyzed reduction of amides to amines.
Angew. Chem., Int. Ed. 2009, 48, 9507−9510.
ORCID
Yusuke Sunada: 0000-0002-8954-181X
Hideo Nagashima: 0000-0001-9495-9666
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Integrated Research
Consortium on Chemical Sciences (IRCCS), the Cooperative
Research Program of “Network Joint Research Center for
Materials and Devices”, and JSPS KAKENHI Grant Numbers
JP17K17944 (A.T.) and JP18K19082 (H.N.).
REFERENCES
■
(1) (a) Marcinec, B. Comprehensive handbook on hydrosilylation;
Pergamon: Oxford, U.K., 1992. (b) Ito, J.-i.; Nishiyama, H.
Bifunctional phebox complexes for asymmetric catalysis. Top.
Organomet. Chem. 2011, 37, 185−205. (c) Larson, G. L.; Fry, J. L.
Ionic and organometallic-catalyzed organosilane reductions. In
Organic Reactions; Denmark, S. E., Ed.; Wiley: New York, 2008;
Vol. 71, Chapter 1, pp 1−737. (d) Marciniec, B. Hydrosilylation: A
Comprehensive Reviews on Recent Advances; Springer: Berlin, 2010.
(2) (a) Nagashima, H. Catalyst design of iron complexes. Bull. Chem.
Soc. Jpn. 2017, 90, 761−775. (b) Nagashima, H.; Sunada, Y.;
Nishikata, T.; Chaiyanurakkul, A. Iron-promoted reduction reactions.
In PATAI’s Chemistry of Functional Groups, The chemistry of
Organoiron Compounds; Patai, S., Ed.; Wiley: Chichester, England,
2014; Chapter 9, pp 325−378.
(8) Pesti, J.; Larson, G. L. Tetramethyldisiloxane: A Practical
Organosilane Reducing Agent. Org. Process Res. Dev. 2016, 20, 1164−
1181.
(9) (a) Motoyama, Y.; Aoki, M.; Takaoka, N.; Aoto, R.; Nagashima,
H. Highly efficient synthesis of aldenamines from carboxamides by
iridium-catalyzed silane-reduction/dehydration under mild condi-
tions. Chem. Commun. 2009, 12, 1574−1576. (b) Tahara, A.;
Miyamoto, Y.; Aoto, R.; Shigeta, K.; Une, Y.; Sunada, Y.;
Motoyama, Y.; Nagashima, H. Catalyst Design of Vaska-Type Iridium
Complexes for Highly Efficient Synthesis of π-Conjugated Enamines.
Organometallics 2015, 34, 4895−4907.
(10) (a) Sinicropi, J. A.; Cowdery-Corvan, J. R.; Magin, E. H.;
Borsenberger, P. M. Hole transport in vapor deposited enamines and
enamine doped polymers. Chem. Phys. 1997, 218, 331−339.
(b) Matoliukstyte, A.; Burbulis, E.; Grazulevicius, J. V.; Gaidelis, V.;
Jancauskas, V. Carbazole-containing enamines as charge transport
materials for electrophotography. Synth. Met. 2008, 158, 462−467.
(c) Paspirgelyte, R.; Grazulevicius, J. V.; Grigalevicius, S.; Jankauskas,
V. Phenoxazine and N-phenyl-1-naphtylamine-based enamines as
hole-transporting glass-forming materials. Synth. Met. 2009, 159,
1014−1018. Related compounds (bisenamines) were reported:
(d) Puodziukynaite, E.; Burbulis, E.; Grazulevicius, J. V.; Jancauskas,
V.; Undezenas, A.; Linonis, V. Aniline-based bis(enamines) as new
amorphous molecular charge transport materials. Synth. Met. 2007,
157, 696−701. (e) Puodziukynaite, E.; Burbulis, E.; Grazulevicius, J.
(3) (a) Addis, D.; Das, S.; Junge, K.; Beller, M. Selective Reduction
of Carboxylic Acid Derivatives by Catalytic Hydrosilylation. Angew.
Chem., Int. Ed. 2011, 50, 6004−6011. (b) Igarashi, M.; Fuchikami, T.
Transition-metal complex-catalyzed reduction of amides with hydro-
silanes: a facile transformation of amides to amines. Tetrahedron Lett.
2001, 42, 1945−1947. (c) Igarashi, M.; Mizuno, R.; Fuchikami, T.
Ruthenium complex catalyzed hydrosilylation of esters: a facile
transformation of esters to alkyl silyl acetals and aldehydes.
Tetrahedron Lett. 2001, 42, 2149−2151. (d) Kuwano, R.;
Takahashi, M.; Ito, Y. Reduction of amides to amines via catalytic
hydrosilylation by a rhodium complex. Tetrahedron Lett. 1998, 39,
1017−1020. (e) Matsubara, K.; Iura, T.; Maki, T.; Nagashima, H. A
I
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
V.; Getausis, V.; Jancauskas, V. Carbazole-based bis(enamines) as
effective charge-transporting amorphous molecular materials. Synth.
Met. 2008, 158, 993−998. (f) Paspirgelyte, R.; Zostautiene, R.; Buika,
G.; Grazulevicius, J. V.; Grigalevicius, S.; Jankauskas, V.; Chen, C.-C.;
Wang, W.-B.; Jou, J. H. Indole and phenylenediamine based enamines
as amorphous hole-transporting materials. Synth. Met. 2010, 160,
162−168.
(11) (a) Greene, J.; Curtis, M. D. Catalysis of siloxane metathesis by
cyclometalladisiloxanes. Mechanistic similarities to olefin metathesis
catalysis. J. Am. Chem. Soc. 1977, 99, 5176−5177. (b) Curtis, M. D.;
Greene, J. Small Ring Metallocycles. 4. Synthesis and Chemical
Reactivity of cyclo-Metalladisiloxanes, MSiR2OSiR2, and cyclo-Metal-
ladisilabutanes, MSiR2CH2SiR2. J. Am. Chem. Soc. 1978, 100, 6362−
6367.
(12) (a) Geoffroy, G. L.; Wrighton, M. S. Organometallic
Photochemistry; Academic Press: 1979. (b) Wink, D. A.; Ford, P. C.
Common intermediates in the flash photolysis of Vaska’s compound
IrCl(CO)(PPh3)2 and its dihydride adduct H2IrCl(CO)(PPh3)2.
Implications with regard to reductive elimination mechanisms. J. Am.
Chem. Soc. 1985, 107, 5566−5567. (c) Wink, D. A.; Ford, P. C.
Reaction dynamics of the tricoordinate intermediates MCl(PPh3)2 (M
= Rh or Ir) as probed by the flash photolysis of the carbonyls
MCl(CO)(PPh3)2. J. Am. Chem. Soc. 1987, 109, 436−442.
(13) Wink and Ford reported the generation of IrCl(PPh3)2 by flush
photolysis of 1a. This 14-electron species is very unstable and reactive.
In the actual catalytic reactions, IrCl(PPh3)2 generated is stabilized by
solvents, amides, or hydrosilanes. In this paper, we describe the
stabilized IrCl(PPh3)2 as “ClIr(PPh3)2”. In a similar fashion,
“HIr(PPh3)2” and “SiIr(PPh3)2” are the species that 14-electron
IrH(PPh3)2 and IrSi(PPh3)2 (Si = −SiMe2OSiMe2H) are stabilized by
solvents, amides, or hydrosilanes.
(14) Bennett, M. A.; Latten, J. L. 35. An Iridium(III) Complex
Containing Cyclometallated Triphenylphosphine formed by Isomer-
ization of an Iridium(I) Triphenylphosphine Complex. Inorg. Synth.
2007, 26, 200−203.
(15) Wilkinson, G. 25. Carbonyl Hydrido Tris(triphenylphosphine)
Iridium (I). Inorg. Synth. 2007, 13, 126−131.
(16) Fernandez, M. J.; Esteruelas, M. A.; Covarrubias, M.; Oro, L. A.
Preparation of IrH(diene)L2 Compounds via Methoxyiridium
Complexes: Catalysts for Hydrogen Transfer Reactions. J. Organomet.
Chem. 1986, 316, 343−349.
(17) Crabtree, R. H.; Felkin, H.; Morris, G. E. Cationic Iridium
Diolefin Complexes as Alkene Hydrogenation Catalysts and the
Isolation of Some Related Hydrido Complexes. J. Organomet. Chem.
1977, 141, 205−215.
Commun. 1973, 629−630. (b) Duckett, S. B.; Newell, C. L.;
Eisenberg, R. Observation of New Intermediates in Hydrogenation
Catalyzed by Wilkinson’s Catalyst, RhCl(PPh3)3, Using Para-
hydrogen-induced Polarization. J. Am. Chem. Soc. 1994, 116,
10548−10556. (c) Goodman, J.; Grushin, V. V.; Larichev, R. B.;
Macgregor, S. A.; Marshall, W. J.; Roe, D. C. Fluxionality of
[(Ph3P)3M(X)] (M = Rh, Ir). The Red and Orange Forms of
[(Ph3P)3Ir(Cl)]. Which Phosphine Dissociates Faster from Wilkin-
son’s Catalyst? J. Am. Chem. Soc. 2010, 132, 12013−12026.
(24) Sunada, Y.; Fujimura, Y.; Nagashima, H. Synergistic Effects of
Two Si-H Groups and a Metal Center” in Transition Metal-Catalyzed
Hydrosilylation of Unsaturated Molecules: A Mechanistic Study of
the RhCl(PPh3)3-Catalyzed Hydrosilylation of Ketones with 1,2-
Bis(dimethylsilyl)benzene. Organometallics 2008, 27, 3502−3513.
(25) (a) Chalk, A. J.; Harrod, J. F. Homogeneous Catalysis. II. The
Mechanism of the Hydrosilation of Olefins Catalyzed by Group VIII
Metal Complexes. J. Am. Chem. Soc. 1965, 87, 16−21. (b) Sakaki, S.;
Mizoe, N.; Sugimoto, M.; Musashi, Y. Pt-catalyzed hydrosilylation of
ethylene. A theoretical study of the reaction mechanism. Coord. Chem.
Rev. 1999, 190−192, 933−960.
(26) (a) Ojima, I.; Nihonyanagi, M.; Nagai, Y. Rhodium complex
catalysed hydrosilylation of carbonyl compounds. J. Chem. Soc., Chem.
Commun. 1972, 938a−938a. (b) Ojima, I.; Nihonyanagi, M.; Kogure,
T.; Kumagai, M.; Horiuchi, S.; Nakatsugawa, K.; Nagai, Y. Reduction
of carbonyl compounds via hydrosilylation: I. Hydrosilylation of
carbonyl compounds catalyzed by tris(triphenylphosphine)-
chlororhodium. J. Organomet. Chem. 1975, 94, 449−461. (c) Sakaki,
S.; Sugimoto, M.; Fukuhara, M.; Sugimoto, M.; Fujimoto, H.;
Matsuzaki, S. Why Does the Rhodium-Catalyzed Hydrosilylation of
Alkenes Take Place through a Modified Chalk-Harrod Mechanism? A
Theoretical Study. Organometallics 2002, 21, 3788−3802. (d) Sakaki,
S.; Sugimoto, M. A Theoretical Study on the Oxidative Addition of a
Si-H σ-Bond to [MCl(CO)(PH3)2] (M = Rh or Ir). Similarities to
and Differences from [M′(PH3)2] (M′ = Pd or Pt) and [RhCl-
(PH3)2]. Bull. Chem. Soc. Jpn. 1996, 69, 3047−3057.
(27) (a) Chalk, A. J.; Harrod, J. F. Dicobalt Octacarbonyl as a
Catalyst for Hydrosilylation of Olefins. J. Am. Chem. Soc. 1965, 87,
1133−1133. (b) Chalk, A. J.; Harrod, J. F. Reactions between
Dicobalt Octacarbonyl and Silicon Hydrides. J. Am. Chem. Soc. 1965,
87, 1133−1135. (c) Chalk, A. J.; Harrod, J. F. Homogeneous
Catalysis. IV. Some Reactions of Silicon Hydrides in the Presence of
Cobalt Carbonyls. J. Am. Chem. Soc. 1967, 89, 1640−1647.
(28) (a) Reichel, C. L.; Wrighton, M. S. Photochemistry of Cobalt
Carbonyl Complexes having a Cobalt-silicon Bond and its Importance
in Activation of Catalysis. Inorg. Chem. 1980, 19, 3858−3860.
(b) Mitchener, J. C.; Wrighton, M. S. Photogeneration of Very Active
Homogeneous Catalysts using Laser Light Excitation of Iron
Carbonyl Precursors. J. Am. Chem. Soc. 1981, 103, 975−977.
(29) In the original report by Chalk and Harrod, insertion of alkenes
into a H−CoI bond of HCo(CO)n was proposed as a plausible
mechanism.27 In contrast, that into the Si-Co(I) species of Si-
Co(CO)n was discussed as an alternative by Wrighton and co-
workers.28 Oxidative addition of a H−Si bond to the resulting ethyl or
silylethyl Co(I) species induces the reductive elimination of the
hydrosilylated product.27,28 Similar mechanisms, namely those
involving insertion of a CO of the amide into either “H-IrI(PPh3)2”
or “Si-Ir(I)(PPh3)2”, should be investigated in further mechanistic
considerations using DFT calculations.
(30) Nagashima, H.; Tatebe, K.; Ishibashi, I.; Nakaoka, A.;
Sakakibara, J.; Itoh, K. Unusual Rate Enhancement in the RhCl-
(PPh3)3-Catalyzed Hydrosilylation by Organosilanes Having Two Si-
H Groups at Appropriate Distances: Mechanistic Aspects. Organo-
metallics 1995, 14, 2868−2879.
(31) Nakatani, N.; Hasegawa, J.; Sunada, Y.; Nagashima, H.
Platinum-catalyzed reduction of amides with hydrosilanes bearing
dual Si−H groups: a theoretical study of the reaction mechanism.
Dalton Trans. 2015, 44, 19344−19356.
(32) When 1a was used as the catalyst, dissociation of CO generated
catalytically active species. In the catalytic reaction, the active species
(18) (a) Casey, C. P.; Rutter, E. W., Jr. Hydrogen Transfer from
CpRe(CO)2H2 to trans-Ir(CO)(Cl)(PPh3)2 via a Heterobimetallic
Re-Ir Intermediate. Inorg. Chem. 1990, 29, 2333−2335. (b) Park, S.;
Lough, A. J.; Morris, R. H. Iridium(III) Complex Containing a
Unique Bifurcated Hydrogen Bond Interaction Involving Ir-H···
H(N)···F-B atoms. Crystal and Molecular Structure of [IrH(η1-
SC5H4NH)(η2-SC5H4N)(PPh3)2](BF4)·0.5C6H6. Inorg. Chem. 1996,
35, 3001−3006.
(19) Loza, M.; Faller, J. W.; Crabtree, R. H. Seven-Coordinate
Iridium(V) Polyhydrides with Chelating Bis(silyl) Ligands. Inorg.
Chem. 1995, 34, 2937−2941.
(20) (a) Osborn, J. A.; Wilkinson, G. 17. Chloro Tris
(Triphenylphosphine) Rhodium (I) (Wilkinson’s Catalyst). Inorg,
Synth. 2007, 28, 77−79. (b) Corriu, R. J. P.; Moreau, J. J. E.; Pataud-
Sat, M. Rhodium complex-catalyzed reactions of o-bis(dimethylsilyl)-
benzene with nitriles. J. Organomet. Chem. 1982, 228, 301−308.
(21) Evans, D.; Osborn, J. A.; Wilkinson, G. 18. trans-Carbonyl
Chloro Bis(Triphenylphosphine) Rhodium and Related Complexes.
Inorg, Synth. 2007, 28, 79−80.
(22) Ahmad, N.; Levison, J. J.; Robinson, S. D.; Uttley, M. F. 19.
Hydrido Phosphine Complexes of Rhodium(I). Inorg, Synth. 2007,
28, 81−82.
(23) (a) Halpern, J.; Wong, C. S. Hydrogenation of tris-
(triphenylphosphine)chlororhodium (I). J. Chem. Soc., Chem.
J
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
reacted with TMDS and amide existing in the solution. Coordination
of CO to form 7 is competitive with that of the amide and does not
occur as long as large amounts of amides exist in the reaction medium.
In other words, progress of the catalytic reaction results in lowering
the amide concentration, which triggers the reaction of CO with the
active species to form 7.
(33) Curtis’s complex 7 as the deactivated species was observed in
the catalysis of IrCl(CO)(PPh3)2. The CO ligand required for the
conversion of X2 to 7 originates from the CO ligand in IrCl(CO)
(PPh3)2. Photoassisted dissociation of the CO ligand in 7 caused it to
react with H2 to form X2, which initiates the catalytic hydrosilylation.
It is known that a small amount of H2 is often formed by the catalytic
reaction of hydrosilanes with moisture, which is a source of H2 to
form X2 from 7.
(34) One of the reviewers requested experiments to suggest the
robustness of catalytically active species. As described in section 1-8-1,
page S-8, in the Supporting Information, we carried out the reaction
of 2a with 3a catalyzed by 8 and attempted isolation of the product by
distillation and subsequent catalyst recovery from the residue;
however, the recycling experiment was hampered due to decom-
position of a part of the catalytically active species during the recovery
process. As a supporting experiment, we carried out five successive
runs of the reaction, in which volatiles were removed in vacuo after
the reaction of each run, 2a and TMDS were added, and performed
the next run was performed (section 1-8-2, page S-8, in the
Supporting Information). The yield of the product at each run was
73−97% (TON = 73000−97000), suggesting that the catalytic
activity of each run was not very different, and no substantial
decomposition of the catalytically active species occurred. The total
reached TON = 475000.
(35) (a) Nakajima, M.; Sato, T.; Chida, N. Iridium-Catalyzed
Chemoselective Reductive Nucleophilic Addition to N-Methoxya-
mides. Org. Lett. 2015, 17, 1696−1699. (b) Katahara, S.; Kobayashi,
S.; Fujita, K.; Matsumoto, T.; Sato, T.; Chida, N. An Iridium-
Catalyzed Reductive Approach to Nitrones from N-Hydroxyamides. J.
Am. Chem. Soc. 2016, 138, 5246−5249. (c) Katahara, S.; Kobayashi,
S.; Fujita, K.; Matsumoto, T.; Sato, T.; Chida, N. Reductive Approach
to Nitrones from N-Siloxyamides and N-Hydroxyamides. Bull. Chem.
́
Soc. Jpn. 2017, 90, 893−904. (d) Fuentes de Arriba, A. L.; Lenci, E.;
Sonawane, M.; Formery, O.; Dixon, D. J. Iridium-Catalyzed Reductive
Strecker Reaction for Late-Stage Amide and Lactam Cyanation.
Angew. Chem., Int. Ed. 2017, 56, 3655−3659. (e) Gammack Yamagata,
A. D.; Dixon, D. J. Enantioselective Construction of the ABCDE
Pentacyclic Core of the Strychnos Alkaloids. Org. Lett. 2017, 19,
1894−1897. (f) Huang, P. Q.; Ou, W.; Han, F. Chemoselective
reductive alkynylation of tertiary amides by Ir and Cu(I) bis-metal
sequential catalysis. Chem. Commun. 2016, 52, 11967−11970.
(36) Vrieze, K.; Collman, J. P.; Sears, C. T., Jr.; Kubota, M. 19.
Trans-Chloro Carbonyl Bis(Triphenylphosphine) Iridium. Inorg.
Synth. 2007, 11, 101−104.
(37) Crabtree, R. H.; Quirk, J. M.; Felkin, H.; Fillebeen-Khan, T. An
Efficient Synthesis of [Ir(cod)Cl]2 and Its Reaction with PMe2Ph to
Give fac-[IrH(PMe2C6H4)(PMe2Ph)3]. Synth. React. Inorg. Met.-Org.
Chem. 1982, 12, 407−413.
(38) Kownacki, I.; Marciniec, B.; Steinberger, H.; Kubicki, M.;
Hoffmann, M.; Ziarko, A.; Szubert, K.; Majchrzak, M.; Rubinsztajn, S.
Effect of triorganophosphites on platinum catalyzed curing of silicon
rubber. Appl. Catal., A 2009, 362, 106−114.
(39) (a) Yu, H.; Gao, B.; Hu, B.; Huang, H. Charge-Transfer
Complex Promoted C−N Bond Activation for Ni-Catalyzed Carbon-
ylation. Org. Lett. 2017, 19, 3520−3523. (b) Obora, Y.; Shimizu, Y.;
Ishii, Y. Intermolecular Oxidative Amination of Olefins with Amines
Catalyzed by the Pd(II)/NPMoV/O2 System. Org. Lett. 2009, 11,
5058−5061.
K
DOI: 10.1021/acs.organomet.8b00835
Organometallics XXXX, XXX, XXX−XXX