Carbon monoxide-driven osmium catalyzed reductive amination harvesting WGSR power

Article abstract of DOI:10.1039/d1cy00695a

Herein, we present the first example of Os-catalyzed efficient reductive amination under water-gas shift reaction conditions. The developed catalytic systems are formedin situin aqueous solutions, employ as small as 0.0625 mol% osmium and are capable of delivering reductive amination products for a broad range of aliphatic and aromatic carbonyl compounds and amines. The scope of the reaction, active catalytic systems, possible limitations of the method and DFT-supported mechanistic considerations are discussed in detail in the manuscript.

Full text of DOI:10.1039/d1cy00695a

Catalysis
Science &
Technology
View Article Online
View Journal | View Issue
PAPER
Carbon monoxide-driven osmium catalyzed
reductive amination harvesting WGSR power†
Cite this: Catal. Sci. Technol., 2021,
11, 4922
a
a
a
Klim O. Biriukov, Mikhail M. Vinogradov,
Oleg I. Afanasyev,
b
a
a
Dmitry V. Vasilyev,a
Alexey A. Tsygankov, Maria Godovikova,
ac
ac
Yulia V. Nelyubina, Dmitry A. Loginov and Denis Chusov
*
Herein, we present the first example of Os-catalyzed efficient reductive amination under water–gas shift
reaction conditions. The developed catalytic systems are formed in situ in aqueous solutions, employ as
small as 0.0625 mol% osmium and are capable of delivering reductive amination products for a broad
range of aliphatic and aromatic carbonyl compounds and amines. The scope of the reaction, active
catalytic systems, possible limitations of the method and DFT-supported mechanistic considerations are
discussed in detail in the manuscript.
Received 19th April 2021,
Accepted 20th May 2021
DOI: 10.1039/d1cy00695a
rsc.li/catalysis
The water–gas shift reaction (WGSR) is a promising
alternative to the abovementioned reducing systems. A half-
Introduction
Development of new catalytic approaches for atom-efficient
build-up of complex molecules is a fundamental goal of the
synthetic community. Amines are a crucial class of organic
compounds with a great variety of applications in multiple
cell reduction potential of E° = −0.53 V for the WGSR allows
the comparison of its reducing power to Ga(0), implying the
1
–8
34
selectivity and mildness of this reducing medium.
Meanwhile, the formal reduction equation implies CO to be
2
9
–15
fields.
challenging.
the selective synthesis of amines, which are widely employed
However, synthesis of amines still remains
the only by-product, which is easily removed from the
reaction mixture. Reactions such as reduction of unsaturated
CX bonds (XC, O, N), hydrocarbonylation, reduction of
nitro groups and heterocycles, reductive alkylation and
allylation were successfully performed under WGSR
1
6,17
Reductive amination (RA) is a potent tool for
1
8–20
in the chemical and pharmaceutical industries.
Developing robust and mild catalytic systems for RA is
therefore of high interest and relevance with multiple
heterogeneous and homogeneous active systems reported to
3
4–40
conditions.
RA is not exempt from this list; however, the
scope of organometallic systems which were applied for RA
under WGSR conditions remains limited mainly to the
19,21–28
date.
Unfortunately, the usage of the most prominent
4
1–45
46
47
reducing agents – H and borohydrides – faces a number of
traditional Rh,
Au, and Co catalysts. The application
2
limitations; while borohydride-based systems are not efficient
of the Os-based complexes, in contrast, remains unravelled,
despite them demonstrating robustness and high activity for
2
9
enough in the case of hindered substrates,
produce
48–55
considerable amount of solid waste and require a thorough
both hydrogenation and transfer hydrogenation.
Recently, we demonstrated the first example of Os-catalyzed
3
0
2
workup to isolate the product, reduction by H using simple
3
0
56
catalysts like Pd/C exhibits low selectivity. The selectivity of
hydrogen-based reducing systems could be increased by the
fine tuning of the catalyst structure thus increasing its
RA, where CO serves as the reducing agent.
When
optimizing the conditions for this RA, we noticed a positive
effect of water additives on the performance of some of the
employed catalytic systems. Encouraged by these
observations, we decided to develop a system capable of
efficient promotion of RA under WGSR conditions, using
water as a solvent and reducing agent (Scheme 1). Herein, we
report the successful completion of this task. To the best of
our knowledge, this is the first example of an efficient Os-
catalyzed RA under WGSR conditions.
1
9,31–33
complexity.
Hence, development of more universal
reducing systems for reductive amination is highly desirable.
a
A. N. Nesmeyanov Institute of Organoelement Compounds of Russian Academy of
Sciences (INEOS RAS), Vavilova St. 28, Moscow, Russian Federation.
E-mail: denis.chusov@gmail.com
b
Forschungszentrum Jülich GmbH, Helmholtz Institute Erlangen-Nürnberg for
Renewable Energy (IEK-11), Egerlandstr. 3, 91058 Erlangen, Germany
c
G. V. Plekhanov Russian University of Economics, 36 Stremyanny Per., Moscow
Results and discussion
1
†
2
17997, Russian Federation
Electronic supplementary information (ESI) available. CCDC 2075331 and
075332. For ESI and crystallographic data in CIF or other electronic format see
For assessing the activity of the Os compounds for RA under
DOI: 10.1039/d1cy00695a
WGSR conditions, p-anisaldehyde and p-anisidine were
4922 | Catal. Sci. Technol., 2021, 11, 4922–4930
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
amounts of the target amine. The simplest Os source Na2-
OsCl6 afforded an almost negligible amount of the desired
product as well. Therefore, we decided to tune the activity of
the catalytic species by introducing bidentate 2,2′-bipyridyl-
like ligands affording complexes Os3–Os8. The synthesized
complexes with one bidentate N,N-ligand and labile CO (Os7
−
−
and Os8), cymene (Os3–Os6), I (Os6) or Cl (Os3–Os5, Os7
and Os8) at the remaining positions were active for RA and
provided good to excellent yields (75–99%). Meanwhile,
complexes with two bidentate ligands Os9 and Os10 were not
active for the catalysis. Last, a Schiff base derived from
p-anisidine and p-anisaldehyde is expected to be formed in
the reaction mixture. In order to assess its potential to serve
as a ligand for the active catalytic species, we synthesized
complexes Os11 and Os12 containing Schiff base-derived C,
N-bidentate ligands. When involved in the reaction mixture
as catalysts, these complexes demonstrated scarce catalytic
performance, which rules them out as possible participants
Scheme 1 Known and proposed systems for RA.
reacted at 120 °C under 50 bar of CO in an aqueous solution
in the presence of Os complexes. Along with sodium
chloroosmate (IV) (Na OsCl ), a total of 12 well-defined
2
6
complexes were tested (Scheme 2). As we previously observed
the moderate activity of Os1 when it was used together with
5
6
silver trifluoroacetate, Os1 and a similar dimeric complex
Os2 were tested for RA under WGSR conditions. Per se, these
compounds were not capable of yielding considerable
Scheme 2 Osmium catalysts screening. Conditions: 2 equiv. p-anisidine, 1 equiv. p-anisaldehyde, 0.5 mol% catalyst, water, 50 bar CO, 120 °C, 22 h.
Detailed conditions are provided in the ESI.†
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 4922–4930 | 4923

View Article Online
Paper
Catalysis Science & Technology
of the catalytic cycle. Therefore, the complexes with one
bidentate N,N-ligand were found to be the most active
catalysts for RA within the tested range.
The structures of compounds Os8 and Os12 were
confirmed using the X-ray diffraction data (Fig. 1 and 2). The
chloride bathophenanthroline complex Os8 has a structure
5
7
similar to the iodide derivative (Bphen)OsI (CO) , where
2
2
both carbonyl ligands are located in trans positions to the
phenanthroline moiety. The Os–C bonds in Os8 (av. 1.884 Å)
are only slightly longer than those in the iodide analogue (av.
1
.878 Å), suggesting a weak effect of the halogen nature on
the binding of the osmium atom to the carbonyl ligands.
Moreover, the length of these bonds is almost independent
of the type of N,N-ligand used. For example, closely related
compounds containing 2,2′-bipyridine, 4,4′-dimethyl-2,2′-
bipyridine, and 2,2′-bipyridine-4,4′-dicarboxylic acid have
5
8,59
Os–C bonds with lengths in the range of 1.873–1.887 Å.
Fig. 2 General view of compound Os12 with atoms shown as thermal
ellipsoids at the 50% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths (Å): Os1–Cl1 2.4134(17), Os1–N1 2.103(5),
Os1–C1 2.187(6), Os1–C2 2.260(6), Os1–C3 2.263(6), Os1–C4 2.214(6),
Os1–C5 2.191(7), Os1–C6 2.160(7), Os1–C11 2.075(6), N1–C17 1.303(8),
C16–C17 1.413(9), C11–C16 1.425(9).
To the best of our knowledge, compound Os12 is the first
example of a cyclometalated osmium complex with a Schiff
base, which is formed via direct CH activation. The structure
contains a five-membered C NOs cycle (Fig. 2). Noteworthily,
3
the formation of this complex is accompanied by
rearrangement of bonds within the N1–C17–C16–C11 moiety
compared
methylaniline.
to
the
free
N-(4-methoxybenzylidene)-4-
6
0
Along with this rearrangement, the
have a rigid structure with a fixed dihedral angle, which
appears to be beneficial for the catalysis; ligands that possess
a more flexible backbone and are prone to not remain
coplanar (bipy L1 and 2,2′-biquinoline L4) afforded inferior
yields (with a still good yield of 89% for L1 and scarce yield
of 5% for L4). Meanwhile, the introduction of steric
hindrance into the vicinity of the N-centers (neocuproine L5)
diminished the performance of the system.
When dimeric complex Os1 was utilized as the Os
precursor along with the bipy ligand, the RA product was
formed with a good yield of 88% (Table 1, entry 13). To assess
how the choice of a solvent impacts the course of catalysis,
some of the most common solvents were tested for RA under
the same conditions (Table 1, entries 12–16).
elongation of the N1–C17 and C16–C11 bonds (1.303 vs.
.274 and 1.425 vs. 1.398 Å, respectively) and the
shortening of the C17–C16 bond (1.413 vs. 1.471 Å) are
observed.
1
Å
In order to test whether the catalytic species can be
formed in situ in the course of the reaction, a series of
experiments utilising Os precursors and bidentate N,N-
ligands were performed (Table 1, entries 2–17, Fig. 3). When
the simplest Na OsCl was employed, bathophenanthroline
2
6
L3 and 1,10-phenanthroline L7 afforded the best yields of the
RA product among all the tested ligands, assuring
quantitative formation of the secondary amine. These ligands
Among the used solvents, water and alcohols (MeOH and
EtOH) had superior performance with yields exceeding 60%,
while the usage of water-free (approx. 20 ppm H O) aprotic
2
solvents yielded ca. 10% of the target product (Table 1,
entries 12 and 14). Additional optimization data including
solvent screening with the indicated water content are
provided in the ESI.† In the case of using water-free solvents,
the reaction could not take place via the WGSR. In this case
the reaction can proceed via the reductive addition without
6
1
an external hydrogen source mechanism.
As the target reaction proceeded smoothly in both water
and ethanol, different mixtures of these two solvents were
tested as reaction media in order to assess the possible
synergistic effect (Table S9†). Interestingly, the 1 : 4 H O–
EtOH and 4 : 1 H O–EtOH mixtures yielded considerably
2
more product compared to the pure water and the 1 : 1
mixture, revealing a complex dependency of the reaction
kinetics on the composition of the solvent. Nevertheless,
2
Fig. 1 General view of compound Os8 with atoms shown as thermal
ellipsoids at the 50% probability level. Hydrogen atoms are omitted for
clarity. Selected bond lengths (Å): Os1–Cl1 2.388(4), Os1–Cl2 2.400(4),
Os1–N1 2.134(11), Os1–N2 2.123(11), Os1–C25 1.884(17), Os1–C26
1
.884(15), O1–C25 1.129(17), O2–C26 1.122(16).
4924 | Catal. Sci. Technol., 2021, 11, 4922–4930
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
Table 1 Selected results for the testing of performance of Os-based systems for RA
a
Entry
Source of Os
Additive
Solvent
Yield, %
1
2
3
4
5
6
7
8
Na
Na
Na
Na
Na
Na
Na
Na
2
2
2
2
2
2
2
2
OsCl
OsCl
OsCl
OsCl
OsCl
OsCl
OsCl
OsCl
6
6
6
6
6
6
6
6
None
L1
L2
L3
L4
L5
L6
L7
L1
L3
L7
L1
L1
L1
L1
L1
L1
H
H
H
H
H
H
H
H
H
H
H
THF
2
H O
MeCN
MeOH
EtOH
2
2
2
2
2
2
2
2
2
2
2
O
O
O
O
O
O
O
O
1
89
4
99
5
4
11
99
96
74
71
12
88
10
62
62
99
b
c
9
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
[(Cymene)OsCl
2
2
2
2
2
2
2
2
]
]
]
]
]
]
]
]
2
2
2
2
2
2
2
2
Os1
Os1
Os1
Os1
Os1
Os1
Os1
Os1
O/EtOHc
b
10
11
12
13
14
15
16
17
O/EtOH
b
b
b
b
b
b
c
O/EtOH
[(Cymene)OsI
2
]
2
Os2
2
H O
a
2
equiv. 2 equiv. p-anisaldehyde, 1 equiv. p-anisidine, 0.5 mol% catalyst, 1.5 mol% additive, solvent, 50 bar CO, 120 °C, 22 h. Detailed
b c
conditions and full optimization data are provided in the ESI.† 0.5 mol% of the ligand was added.
2
H O/EtOH = 4/1.
adding 20 vol% of EtOH to the aqueous reaction mixture
appears to be a simple and powerful way to improve the
reaction ability of the system.
range of synthesized Os complexes (Fig. 4) in the presence or
absence of bipy (L1) as the ligand. Among all the tested
catalytic mixtures, the combination of Os2 and L1 was
performing in an exceptional way, affording the RA product
in moderate or good yields even with a loading as low as
0.0625%. However, to ensure highest possible yields, a
loading of 0.25% can be considered as optimal. Considering
the combination of the ligand price, ease of preparation and
efficient metal loading, the Os2/bipy system is of the highest
interest for practical application (Table S13†).
The same range of ligands as those for the Na OsCl /water
2
6
2 2
system was tested for the discovered [(Cymene)OsCl ] /EtOH/
H
2
O system. The most successful systems are shown in
Table 1, entries 9–11. Similar to the Na OsCl /H O system,
2
6
2
the ligands bipy, bathophenanthroline and 1,10-
phenanthroline showed the best performance. Finally,
[
(Cymene)OsI ] was used as the source of Os (Table 1, entry
2 2
1
7). When used in water with bipy as the ligand, the resulting
To get a better understanding of the reactivity of this reducing
system, some mechanistic investigations were carried out.
system with 99% yield of the RA product significantly
outperformed Na OsCl and [(Cymene)OsCl2]2 with 89 and
8
During the optimization, we found that the excess of bipy
(L1) has a positive effect on the yield of the reaction with an
optimal loading of 2–3 equivalents with respect to the Os
source (Table S4†). As complexes with more than one N,N-
ligand have low catalytic activity (see above), we rationalize
the beneficial effect of the ligand excess either by a faster
kinetics of the formation of catalytic species or by facilitating
the formation of hydroxyl anions from water, which is an
important step in the conventional mechanism for the WGSR-
2
6
8% yields, respectively.
Further optimization of the RA conditions revealed that
two-fold excess of amine, 50 bar CO and 120–140 °C resulted
in the highest yields of the products. A decrease in any of
these parameters lowered the performance of the systems.
Detailed optimization tables are provided in the ESI.†
As the last optimization step, the effect of the catalyst
loading on the activity of the systems was assessed for a
3
4
driven catalytic reductive amination (Scheme 4, step I).
Additionally, we checked which osmium species are
formed in the reaction mixture in the presence and in the
absence of carbon monoxide and substrates (Scheme 3, a–c).
Under argon atmosphere heating Os2 with bipy ligand gives
the corresponding (Cymene)Os(bipy)I
The same reaction proceeds under carbon monoxide
atmosphere, accompanied by the formation of
2
species (Scheme 3, a).
a
sophisticated mixture of osmium species. However, during
this experiment a complex mixture of other identified
Fig. 3 Ligands utilized for the in situ generation of catalysts.
products was formed. Compound (bipy)
2
Os(CO)I
2
was
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 4922–4930 | 4925

View Article Online
Paper
Catalysis Science & Technology
Fig. 4 Effect of the loading of the Os catalyst on the RA yield. Conditions: 2 equiv. p-anisidine, 1 equiv. p-anisaldehyde, x mol% catalyst, solvent,
0 bar CO, 120 °C, 22 h.
5
identified among them using mass spectrometry
Scheme 3, b). The reaction in the high pressure NMR tube
reductant in the RA cycle. The second variant implies the in
situ formation of a metal hydride or alternative species with
high reducing ability without intermediate formation of
dihydrogen (Scheme 4).
We tested whether hydrogen can serve as a reducing agent
for the relevant Schiff base (Scheme 3, d) in THF and water
as reaction media. The reaction in absolute THF resulted in
the formation of the target product with a high yield to
(
revealed the formation of (Cymene)Os(bipy)I (Scheme 3, c).
2
Finally, we tried to elucidate the origin of the actual
reducing agent in the system. Generally, two mechanisms of
3
4
the WGSR-driven reduction are possible. The first implies
the formation of hydrogen from carbon monoxide and water
in one catalytic cycle, with the evolved hydrogen serving as a
Scheme 3 Mechanistic studies to reveal the catalytically active species and real reductant. a) Reaction of Os2 with bipy ligand in the absence of
CO; b) Reaction of Os2 with bipy ligand in the presence of CO; c) Reaction with equimolar amounts of osmium and bipy in the high-pressure NMR
tube; d) Reduction of Schiff base with hydrogen in water and water-free medium.
4926 | Catal. Sci. Technol., 2021, 11, 4922–4930
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
Scheme
reductive amination.
4 Proposed mechanism of the catalytic WGSR-driven
3
4
Scheme
5
Substrate scope for the osmium-catalyzed reductive
conversion ratio (42% yield, 56% conversion). When water
was used as the solvent, the conversion was lower, and
various by-products (e.g. Schiff base hydrolysis) were
detected. This indicates that hydrogen is not the actual
reductant, and the reaction takes place in one cycle in a
manner depicted in Scheme 4.
The optimized RA conditions were employed to test the
scope of the established method. The synthesized products
and their yields are shown in Scheme 5. The developed
protocol allows us to work with different combinations of
aromatic and aliphatic amines, aldehydes and ketones. Even
aliphatic aldehydes, which might proceed with the self-aldol
reaction in the presence of amines, undergo the reductive
amination reaction. The presence of ortho-substituents in the
aldehyde structure (3) does not alter the reaction. Aromatic
chloride (4 and 5), O-benzyl (6) and N-benzyl moieties in
a
amination. NMR yields. Isolated yields are indicated in parentheses. 140
b
c
d
e
f
°
C, 160 °C, acetone 20 equiv., 180 °C, 0.0625 mol% Os2, 5 equiv.
amine.
unknown products. This is the intrinsic problem of the
reductive amination with an external hydrogen source in the
presence of reducible functional groups under harsh
conditions. This could be avoided by using RA without an
external hydrogen source
catalytic systems allowing the use of milder conditions.
Meanwhile, an attempt to introduce acetophenone in RA with
morpholine yielded only 17% of the desired amine, while
1-phenylethanol was the predominant product (55–75%)
6
1,63
or developing more active
3
8
(Scheme 6, c).
A similar situation was noted when
benzophenone was utilized as the carbonyl component
(Scheme 6, d). In this case, only diphenylmethanol was
detected with no evidence for the target product formation.
However, in the case of the reaction between acetophenone
and p-anisidine, the amine was the main reduction product
with no 1-phenylethanol detected (Scheme 6, e). This
indicates that ketone reduction takes place in the presence of
aliphatic amines with high basicity. Aromatic amines are
not strong enough in terms of basicity to promote this
reaction. We rationalize the formation of alcohols by an
increase of the reducing potential of the WGSR-based system
in the presence of promoters possessing high basicity.
Further development of other catalytic systems might
overcome the revealed selectivity problems.
To elucidate the origin of the positive impact of N,N-
ligands on the Os-catalysed RA, we performed DFT
calculations at the M06-L/SDD level. Taking into account that
water was proven to be the best solvent, we proposed that the
products (1–6), which can be easily reduced by H
2
over Pd/C
3
0,62
at room temperature,
are well tolerated as well. The
system was capable of converting morpholine to the RA
product even at 0.0625% catalyst loading with 80% yield (14).
However, attempts to conduct RA using aldehydes with
other reducible functional groups or less electrophilic
aromatic ketones were unsuccessful (Scheme 6). For example,
p-cyanobenzaldehyde afforded only 6% of the RA product
along with 76% of the corresponding Schiff base with 100%
conversion at 120 °C (Scheme 6, a). Increasing the
temperature allowed 32% of the target molecule to be
obtained; however, the reaction became more sluggish. A
further increase in temperature led to full degradation of the
cyano-group with the formation of a complex mixture of
unidentified products. A similar trend was observed for
m-nitrobenzaldehyde (Scheme 6, b): at low temperature, only
the Schiff base was formed, while at elevated temperatures
the nitro group was reduced resulting in a cocktail of
3
8
6
3,64
3
4,65
reaction proceeds through the WGSR path.
Although this
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 4922–4930 | 4927

View Article Online
Paper
Catalysis Science & Technology
−1
Table 2 Activation barriers (in kcal mol ) for the decarboxylation step in
different proposed intermediates. (See step II in Scheme 3)
‡
Entry
1
Transition state
ΔG
58.1
2
3
53.8
39.5
4
38.2
Scheme 6 Challenging examples of the reductive amination under
WGSR conditions. The scheme illustrates following problems: a)
Reduction of the cyano group; b) Reduction of nitro group; c) and d)
Reduction of aromatic ketone in the presence of aliphatic amine; e)
Low conversion of aromatic ketone in the presence of aromatic amine.
5
38.7
mechanism has been computationally studied in detail for
6
6–69
neutral group 8 pentacarbonyls,
the effect of the ligand
nature on the activation barriers has not been estimated
before. Recently, Smith and co-workers demonstrated that
the decarboxylation with the formation of hydride is the rate
determining step for the WGSR catalysed by ruthenium(II)
7
0
complexes (step II, Scheme 3). We calculated the barriers of
this step for different proposed Os intermediates (Table 2).
In the absence of an N,N-ligand, the octahedral complex
ability of the chloride-anion compared to that of the carbonyl
ligand.
Os(CO)
4
Cl
2
is the most probable starting catalytic species,
Probably, complex Os7 is a key catalytic species in all the
reactions when the bipy ligand was present in the system.
This complex on its own shows high efficiency for RA in
which can be formed as a result of elimination of the cymene
ligand from [(Cymene)OsCl
7
1
2
]
2
.
Hydroxylation of one of its
carbonyl ligands can give two isomeric intermediates with
different positions of the carboxylic group with respect to
other ligands: either opposite to the carbonyl (Table 2, entry
water (Scheme 2). We found that the introduction of the bipy
‡
ligand results in
a
crucial decrease of ΔG for the
−
1
decarboxylation step to 39.5 kcal mol (Table 2, entry 3),
which correlates well with the experimentally observed high
activity of the complex for RA. Notably, the inclusion of a
water molecule into the transition state, as well as the
replacement of chloride with iodide ligands, leads to a
1
) or opposite to the chloride moiety (Table 2, entry 2). The
‡
activation barrier (ΔG ) of the subsequent decarboxylation is
5
−
1
8.1 kcal mol for the former (trans-CO) isomer. At the same
time, in the case of the second (trans-Cl) isomer, the
−
1
−1
calculated barrier is smaller by 4.3 kcal mol . The obtained
result can be rationalized by the stronger electron donating
further decrease of the barrier by less than 2 kcal mol
(Table 2, entries 4 and 5 vs. entry 3). Thus, we conclude that
4928 | Catal. Sci. Technol., 2021, 11, 4922–4930
This journal is © The Royal Society of Chemistry 2021

View Article Online
Catalysis Science & Technology
Paper
the trans-effect of incoming ligands (such as bipy and L7) can
considerably facilitate the decarboxylation step in the WGSR
with the formation of Os hydrides.
equipment of the Center for molecular composition studies
of INEOS RAS. D. A. L. and D. C. are also thankful to the
Plekhanov Russian University of Economics for providing
access to computation resources for DFT calculations.
Experimental
Notes and references
General procedure for the reductive amination: A glass vial in
a 10 mL stainless steel or titanium autoclave was charged
with a prescribed quantity of the catalyst, co-catalyst, amine
1
2
M. von Korff and T. Sander, Sci. Rep., 2019, 9, 967.
L. Marin, G. Force, R. Guillot, V. Gandon, E. Schulz and D.
Lebœuf, Chem. Commun., 2019, 55, 5443–5446.
(50–200 mol%, 0.1–0.4 mmol), carbonyl compound (100
mol%, 0.4 mmol) and 0.4 mL of the corresponding solvent if
mentioned. The autoclave was sealed, flushed three times
with 10 bar of CO, and then charged with CO under the
indicated pressure. The reactor was placed into a preheated
oil bath. After the indicated time, the reactor was cooled to
room temperature and depressurized. The reaction mixture
was transferred into a measuring flask and diluted with
dichloromethane to 5 mL, and then it was analysed by GC or
NMR. Isolation was performed by the means of column
chromatography. All other experimental details including
compound characterization, experimental procedures, full
optimization data and copies of NMR spectra are provided in
the ESI.†
3
4
5
O. Méndez-Lucio and J. L. Medina-Franco, Drug Discovery
Today, 2017, 22, 120–126.
V. Canale, V. Frisi, X. Bantreil, F. Lamaty and P. Zajdel,
J. Org. Chem., 2020, 85, 10958–10965.
N. Marinus, N. Tahiri, M. Duca, L. M. C. M. Mouthaan, S.
Bianca, M. van den Noort, B. Poolman, M. D. Witte and A. J.
Minnaard, Org. Lett., 2020, 22, 5622–5626.
A. Labattut, S. Abi Fayssal, J. Buendia, I. Abdellah, V. Huc,
C. Martini and E. Schulz, React. Chem. Eng., 2020, 5,
6
7
1
509–1514.
J. P. Berndt, Y. Radchenko, J. Becker, C. Logemann, D. R.
Bhandari, R. Hrdina and P. R. Schreiner, Chem. Sci.,
2
019, 10, 3324–3329.
S. Mkrtchyan and V. O. Iaroshenko, Chem. Commun.,
020, 56, 2606–2609.
8
9
Conclusions
2
O. I. Afanasyev, E. A. Kuchuk, K. M. Muratov, G. L. Denisov
An efficient Os-based catalytic system was first established
for RA under WGSR conditions. The developed approach
profits from using CO as the reducing agent, in situ
formation of the catalytic system from simple commercially
available compounds, usage of water or water–ethanol
mixtures as a solvent and possibility to employ low Os
precatalyst loadings. Along with optimization of the reaction
conditions, a search for a catalytically active complex was
performed by synthesizing a range of Os complexes and
assessing their activities as catalysts. Complexes with one
bidentate N,N-ligand and labile ligands at the remaining
positions exhibited the highest catalytic activity, implying
these species to be responsible for the catalysis. On the basis
of the obtained experimental results in combination with
DFT studies the role of the bidentate dinitrogen ligand was
explained. We hope that our results will enable the Os
complexes to find their niche among efficient noble metal
catalysts for RA as well as other reductive alkylations and
encourage future research on this topic.
and D. Chusov, Eur. J. Org. Chem., 2021, 2021, 543–586.
1
0 P. Roose, K. Eller, E. Henkes, R. Rossbacher and H. Höke, in
Ullmann's Encyclopedia of Industrial Chemistry, Wiley-VCH,
Weinheim, 2015.
1 Y. Shirota, J. Mater. Chem., 2005, 15, 75–93.
2 R. Kumar, N. J. Flodén, W. G. Whitehurst and M. J. Gaunt,
Nature, 2020, 581, 415–420.
1
1
1
1
1
3 J. Hannedouche and E. Schulz, Organometallics, 2018, 37,
4
313–4326.
4 F. Ferretti, D. R. Ramadan and F. Ragaini, ChemCatChem,
019, 11, 4450–4488.
2
5 R. Hrdina, F. M. Metz, M. Larrosa, J.-P. Berndt, Y. Y.
Zhygadlo, S. Becker and J. Becker, Eur. J. Org. Chem.,
2
015, 2015, 6231–6236.
1
1
1
6 T. Schönauer, S. L. J. Thomä, L. Kaiser, M. Zobel and R.
Kempe, Chem. – Eur. J., 2021, 27, 1609–1614.
7 K. Banert, M. Heck, A. Ihle, J. Kronawitt, T. Pester and T.
Shoker, J. Org. Chem., 2018, 83, 5138–5148.
8 O. I. Afanasyev, E. Kuchuk, D. L. Usanov and D. Chusov,
Chem. Rev., 2019, 119, 11857–11911.
9 T. Irrgang and R. Kempe, Chem. Rev., 2020, 120, 9583–9674.
0 K. Murugesan, T. Senthamarai, V. G. Chandrashekhar, K.
Natte, P. C. J. Kamer, M. Beller and R. V. Jagadeesh, Chem.
Soc. Rev., 2020, 49, 6273–6328.
Conflicts of interest
1
2
There are no conflicts to declare.
Acknowledgements
2
1 R. Kempe, M. Klarner, P. Blach, H. Wittkaemper, N. de Jong
and C. Papp, ChemCatChem, DOI: 10.1002/cctc.202100562.
This work was supported by the Russian Science Foundation
(
Grant No. 20-73-00010). Characterization of compounds
22 A. R. Petersen, J. M. V. Lauridsen and J.-W. Lee, Eur. J. Org.
Chem., 2020, 2020, 7368–7372.
23 J. Bianga, N. Kopplin, J. Hülsmann, D. Vogt and T.
Seidensticker, Adv. Synth. Catal., 2020, 362, 4415–4424.
using NMR spectroscopy and X-ray diffraction was performed
with the financial support from the Ministry of Science and
Higher Education of the Russian Federation using the
This journal is © The Royal Society of Chemistry 2021
Catal. Sci. Technol., 2021, 11, 4922–4930 | 4929

View Article Online
Paper
Catalysis Science & Technology
2
2
2
4 P. Yao, J. R. Marshall, Z. Xu, J. Lim, S. J. Charnock, D. Zhu
and N. J. Turner, Angew. Chem., Int. Ed., 2021, 60, 8717–8721.
5 L. an Hu, Y. Zhang, Q. W. Zhang, Q. Yin and X. Zhang,
Angew. Chem., Int. Ed., 2020, 59, 5321–5325.
6 E. A. Savicheva, G. Y. Mitronova, L. Thomas, M. J. Böhm, J.
Seikowski, V. N. Belov and S. W. Hell, Angew. Chem., Int. Ed.,
49 C. Barbato, S. Baldino, M. Ballico, R. Figliolia, S. Magnolia,
K. Siega, E. Herdtweck, P. Strazzolini, G. Chelucci and W.
Baratta, Organometallics, 2018, 37, 65–77.
50 B. Y. Park, T. Luong, H. Sato and M. J. Krische, J. Org. Chem.,
2016, 81, 8585–8594.
51 E. Bayón Castañón, M. Kaposi, R. M. Reich and F. E. Kühn,
Dalton Trans., 2018, 47, 2318–2329.
52 J. C. Babón, M. A. Esteruelas, A. M. López and E. Oñate,
Organometallics, 2020, 39, 2177–2188.
53 M. L. Buil, M. A. Esteruelas, J. Herrero, S. Izquierdo, I. M.
Pastor and M. Yus, ACS Catal., 2013, 3, 2072–2075.
54 A. Acosta-Ramirez, M. Bertoli, D. G. Gusev and M. Schlaf,
Green Chem., 2012, 14, 1178–1188.
55 D. G. Gusev, Organometallics, 2020, 39, 258–270.
56 M. M. Vinogradov, O. I. Afanasyev, Y. V. Nelyubina, G. L.
Denisov, D. A. Loginov and D. Chusov, Mol. Catal.,
2020, 498, 111260.
57 Y. L. Chen, S. W. Lee, Y. Chi, K. C. Hwang, S. B. Kumar,
Y. H. Hu, Y. M. Cheng, P. T. Chou, S. M. Peng, G. H. Lee,
S. J. Yeh and C. T. Chen, Inorg. Chem., 2005, 44,
4287–4294.
58 M. Jakonen, P. Hirva, M. Haukka, S. Chardon-Noblat, F.
Lafolet, J. Chauvin and A. Deronzier, Dalton Trans.,
2007, 3314–3324.
59 J. Jänis, M. Jakonen, L. Oresmaa, P. Hirva, E. Laurila, L.
Vlasova, P. Vainiotalo and M. Haukka, Organometallics,
2010, 29, 1070–1078.
60 J. Harada, M. Harakawa and K. Ogawa, Acta Crystallogr., Sect.
B: Struct. Sci., 2004, 60, 578–588.
61 D. Chusov and B. List, Angew. Chem., Int. Ed., 2014, 53,
5199–5201.
62 I. D. Entwistle and W. W. Wood, in Comprehensive Organic
Synthesis, ed. B. M. Trost and I. B. T.-C. O. S. Fleming,
Pergamon, Oxford, 1991, pp. 955–981.
2
020, 59, 5505–5509.
7 G. Hahn, P. Kunnas, N. de Jonge and R. Kempe, Nat. Catal.,
019, 2, 71–77.
8 C. Bäumler, C. Bauer and R. Kempe, ChemSusChem,
020, 13, 3110–3114.
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
2
2
9 A. F. Abdel-Magid, K. G. Carson, B. D. Harris, C. A. Maryanoff
and R. D. Shah, J. Org. Chem., 1996, 61, 3849–3862.
0 E. Podyacheva, O. I. Afanasyev, A. A. Tsygankov, M.
Makarova and D. Chusov, Synthesis, 2019, 51, 2667–2677.
1 V. I. Tararov, R. Kadyrov, T. H. Riermeier and A. Borner,
Chem. Commun., 2000, 1867–1868.
2 T. Gross, A. M. Seayad, M. Ahmad and M. Beller, Org. Lett.,
2
002, 4, 2055–2058.
3 J. Gallardo-Donaire, M. Ernst, O. Trapp and T. Schaub, Adv.
Synth. Catal., 2016, 358, 358–363.
4 A. Ambrosi and S. E. Denmark, Angew. Chem., Int. Ed.,
2
016, 55, 12164–12189.
5 P. Escaffre, A. Thorez and P. Kalck, J. Mol. Catal., 1985, 33,
7–118.
8
6 S. E. Denmark, Z. D. Matesich, S. T. Nguyen and S. Milicevic
Sephton, J. Org. Chem., 2018, 83, 23–48.
7 S. E. Denmark and Z. D. Matesich, J. Org. Chem., 2014, 79,
5
970–5986.
8 S. E. Denmark, M. Y. S. Ibrahim and A. Ambrosi, ACS Catal.,
017, 7, 613–630.
9 P. Zhou, C. Yu, L. Jiang, K. Lv and Z. Zhang, J. Catal.,
017, 352, 264–273.
2
2
0 H. Q. Li, X. Liu, Q. Zhang, S. S. Li, Y. M. Liu, H. Y. He and Y.
Cao, Chem. Commun., 2015, 51, 11217–11220.
63 O. I. Afanasyev, A. A. Tsygankov, D. L. Usanov, D. S.
Perekalin, N. V. Shvydkiy, V. I. Maleev, A. R. Kudinov and D.
Chusov, ACS Catal., 2016, 6, 2043–2046.
1 Y. Watanabe, M. Yamamoto, T. Mitsudo and Y. Takegami,
Tetrahedron Lett., 1978, 19, 1289–1290.
4
4
2 J. W. Park and Y. K. Chung, ACS Catal., 2015, 5, 4846–4850.
3 V. B. Kharitonov, V. S. Ostrovskii, Y. V. Nelyubina, D. V.
Muratov, D. Chusov and D. A. Loginov, J. Organomet. Chem.,
64 S. A. Runikhina, O. I. Afanasyev, K. Biriukov, D. S. Perekalin,
M. Klussmann and D. Chusov, Chem. – Eur. J., 2019, 25,
16225–16229.
2
020, 925, 121468.
65 R. M. Laine and E. J. Crawford, J. Mol. Catal., 1988, 44,
357–387.
66 Y. Chen, F. Zhang, C. Xu, J. Gao, D. Zhai and Z. Zhao,
J. Phys. Chem. A, 2012, 116, 2529–2535.
67 M. Torrent, M. Solà and G. Frenking, Organometallics,
1999, 18, 2801–2812.
68 H. Schulz, A. Görling and W. Hieringer, Inorg. Chem.,
2013, 52, 4786–4794.
4
4
4 S. A. Runikhina, M. A. Arsenov, V. B. Kharitonov, E. R.
Sovdagarova, O. Chusova, Y. V. Nelyubina, G. L. Denisov,
D. L. Usanov, D. Chusov and D. A. Loginov, J. Organomet.
Chem., 2018, 867, 106–112.
5 V. B. Kharitonov, M. Makarova, M. A. Arsenov, Y. V. Nelyubina,
O. Chusova, A. S. Peregudov, S. S. Zlotskii, D. Chusov and
D. A. Loginov, Organometallics, 2018, 37, 2553–2562.
4
4
4
6 M.-M. Zhu, L. Tao, Q. Zhang, J. Dong, Y.-M. Liu, H.-Y. He
and Y. Cao, Green Chem., 2017, 19, 3880–3887.
7 Y. Sugi, A. Matsuda, K. Bando and K. Murata, Chem. Lett.,
69 N. Liu, L. Guo, Z. Cao, W. Li, X. Zheng, Y. Shi, J. Guo and Y.
Xi, J. Phys. Chem. A, 2016, 120, 2408–2419.
70 R. Stepić, C. R. Wick, V. Strobel, D. Berger, N. Vučemilović-
Alagić, M. Haumann, P. Wasserscheid, A.-S. Smith and D. M.
Smith, Angew. Chem., Int. Ed., 2019, 58, 741–745.
71 A. F. Hill, Angew. Chem., Int. Ed., 2000, 39, 130–133.
1
979, 8, 363–364.
8 G. Chelucci, S. Baldino and W. Baratta, Acc. Chem. Res.,
015, 48, 363–379.
2
4930 | Catal. Sci. Technol., 2021, 11, 4922–4930
This journal is © The Royal Society of Chemistry 2021