Intermediate formation enabled regioselective access to amide in the Pd-catalyzed reductive aminocarbonylation of olefin with nitroarene

Article abstract of DOI:10.1016/S1872-2067(20)63561-6

An efficient route for the palladium-catalyzed reductive aminocarbonylation of olefins with nitroarenes was developed using carbon monoxide (CO) as both reductant and carbonyl source, which enables facile access to amides with excellent regioselectivity a

Full text of DOI:10.1016/S1872-2067(20)63561-6

Chinese Journal of Catalysis 41 (2020) 1152–1160
a v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c h n j c
Article
Intermediate formation enabled regioselective access to amide
in the Pd-catalyzed reductive aminocarbonylation of olefin with
nitroarene
Li Yang, Lijun Shi, Chungu Xia *, Fuwei Li #
State Key Laboratory for Oxo Synthesis and Selective Oxidation, Suzhou Research Institute of LICP, Lanzhou Institute of Chemical Physics (LICP), Chinese
Academy of Sciences, Lanzhou 730000, Gansu, China
A R T I C L E I N F O
A B S T R A C T
Article history:
An efficient route for the palladium-catalyzed reductive aminocarbonylation of olefins with ni-
troarenes was developed using carbon monoxide (CO) as both reductant and carbonyl source,
which enables facile access to amides with excellent regioselectivity and broad substrate scope. It is
found that the counter anions of the Pd catalyst precursors significantly affect the reaction
chemoselectivity and amide regioselectivity. Branched amides were mainly obtained with K2PdCl4
as the metal catalyst, and phosphine ligands had no influence on the regioselectivity but affected the
catalytic reactivity. However, phosphine ligands had significant effects on aminocarbonylation regi-
oselectivity when Pd(CH3CN)4(OTf)2 was used; monodentate phosphines tended to form branched
amides, and bidentate phosphines mainly formed linear amides. Trapping experiments, primary
kinetic studies, and control reactions with all possible N-species reduced from nitroarene indicated
that the catalytic synthesis of branched and linear amides produced nitrene (further converted to
enamide) and aniline, respectively, different from the previous ligand-controlled regioselective
synthesis of amides via the aminocarbonylation of olefins with amines. Furthermore, the proposed
synthesis route could be applied in the synthesis of gram-scale propanil under mild conditions.
© 2020, Dalian Institute of Chemical Physics, Chinese Academy of Sciences.
Received 22 October 2019
Accepted 1 December 2019
Published 5 July 2020
Keywords:
Nitroarenes
Amides
N-intermediate
Aminocarbonylation
Regioselectivity
Published by Elsevier B.V. All rights reserved.
1. Introduction
mation control.
Amides constitute an important class of N-containing car-
Nitroarenes are generally cheaper, more readily available,
and more functionally compatible than anilines as nitrogen
sources [1–4]. Therefore, using nitroarenes as starting mole-
cules is advantageous for preparing nitrogen-containing func-
tional molecules in proper cascade reactions [5–19]. Notably,
nitroarenes can be reduced to several corresponding N-species
under different conditions [5–9,19], which provide various
synthesis routes from nitroarenes via in situ intermediate for-
bonyl compounds. They are ubiquitous in natural products,
functional materials, marketed drugs, and pesticides [20–23],
and are attractive intermediates for diverse synthesis [24,25].
Different from the traditional preparation of amides via con-
densation between carboxylic derivatives and amines, the
aminocarbonylations of olefins with amines have been devel-
oped as amide synthesis routes over the last two decades
[22–34]. However, the reductive carbonylation of olefins with
* Corresponding author. E-mail: cgxia@licp.cas.cn
# Corresponding author. E-mail: fuweili@licp.cas.cn
This work was funded by the National Key R&D Program of China (2018YFB1501600), the Natural Science Foundation of China (21773271,
21972151).
DOI: 10.1016/S1872-2067(20)63561-6 | http://www.sciencedirect.com/science/journal/18722067 | Chin. J. Catal., Vol. 41, No. 7, July 2020

Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
1153
nitroarenes to produce amides has rarely been investigated,
2. Experimental
although it could be a promising alternative with readily avail-
able nitroarenes as the N-resource and better step-economy.
In 2013, Beller and coworkers [35] reported a pioneering
Pd-catalyzed aminocarbonylation of olefins with nitroarenes
(Scheme 1(a)); in their study, the nitroarene was in situ re-
duced to aniline with external H2 as a reducing agent, and linear
amide was obtained as the major product without regioselec-
tivity control [35]. Recently, an interesting procedure of ami-
nocarbonylation of styrene with nitroarene for the synthesis of
α,β-unsaturated amide, was successfully developed by Wu’s
group (Scheme 1(b)). In their study, organometallic Mo(CO)6
was used as a CO source [36]. Moreover, the reductive car-
bonylation of nitroarenes with other partners to produce am-
ides has also rarely been reported. Driver et al. [37] demon-
strated the Pd-catalyzed aminocarbonylation of nitroarenes
with aryl C-H bond, but only substrates with a pyridine direct-
ing group lead to the production of amides. The Ni-catalyzed
reductive aminocarbonylation of nitroarenes with aryl halides
was developed; however, an excess solid reducing agent
(Zn/Mg powder) was needed [38]. Furthermore, these two
examples used expensive Mo(CO)6 as the CO source and
showed a low atom economy. Therefore, it is highly interesting
and desirable to develop new and green catalytic procedures
for the reductive carbonylation of nitroarenes.
It is generally accepted that in the carbonylation (including
aminocarbonylation) of olefin, the phosphine ligand plays a key
role in deciding the regioselectivity. Owing to the diverse
N-species originating from the reduction of NO2 group with CO
[16,17], we expect that a new reductive aminocarbonylation of
olefins with nitroarenes to control regioselectivity could be
developed via the formation of different intermediates. In this
paper, we propose a method for the Pd-catalyzed selective
aminocarbonylation of olefins with nitroarenes using CO as
both reductant and carbonyl source (Scheme 1(c)). Detailed
investigations revealed that the catalytic synthesis of branched
and linear amides produced nitrene (further converted to en-
amide) and aniline, respectively, due to the nature of the Pd salt
catalysts and their counter anions. This is different from the
ligand-controlled regioselectivity in the aminocarbonylation of
olefins with amines. Moreover, this catalytic procedure shows
good olefin and nitroarene substrate scope.
2.1. General methods
NMR spectra were recorded at room temperature at 400
and 101 MHz for 1H and 13C, respectively, with CDCl3, DMSO-d6,
or THF-d8 as the solvent. High resolution mass spectra (HRMS)
were recorded with a Thermo Fisher Scientific LTQ FT Ultra in
DART Positive Mode or an Agilent 6530 Accurate-Mass Q-TOF
LC/MS in ESI mode. IR spectra were recorded with a Nicolet
6700 at room temperature. Melting points were recorded with
a Barnstead Electrothermal IA9100. All non-aqueous reactions
and manipulations were performed using standard Schlenk
techniques. All solvents were purified using a Pure Solv 7-SDS
solvent drying system. Unless otherwise stated, materials were
obtained from commercial suppliers and used without further
purification. All reactions were monitored via thin-layer chro-
matography (TLC) with 0.25 mm silica gel glass plates. A GC-MS
analysis was performed with an Agilent 7890A/5975C GC-MS
system.
2.2. General procedure for the Pd-catalyzed reductive
aminocarbonylation of olefin with nitroarene
2.2.1. General catalytic procedure for the synthesis of branched
amides (3)
Olefin (1, 0.2 mmol), nitroarene (2, 0.2 mmol), K2PdCl4 (3.3
mg, 5 mol%), dppf (5.5 mg, 5 mol%), B(OH)3 (24.8 mg, 0.4
mmol), and THF (2.5 mL) were sequentially added into an au-
toclave, which was then purged three times before charging
with 3.5 MPa CO. The autoclave was put into the heating jacket.
The solution was first stirred at 80 °C for 20 h and then cooled
to room temperature; CO was released carefully once the reac-
tion finished. Then, the solvent was removed under reduced
pressure, and the residue was purified by column chromatog-
raphy on silica gel using pentane/ethyl acetate (4:1) as an elu-
ent to obtain the desired branched product 3.
2.2.2. General catalytic procedure for the synthesis of linear
products (4)
Olefin (1, 0.2 mmol), nitroarene (2, 0.2 mmol),
Pd(CH3CN)4(OTf)2 (5.7 mg, 5 mol%), Xantphos (5.8 mg, 5
mol%), B(OH)3 (24.8 mg, 0.4 mmol), and THF (2.5 mL) were
added into a glass tube, which was then placed in an autoclave.
Then, the autoclave was purged three times and charged with
0.1 MPa CO. The autoclave was put into the heating jacket, and
the solution was stirred at 80 °C for 20 h. Finally, branched
amide 4 was obtained following the previous procedure.
3. Results and discussion
3.1. Reaction conditions optimization
Initially, the reductive aminocarbonylation of styrene (1a)
with 3-chloro-nitrobenzene (2a) was taken as the model reac-
tion. As shown in Table 1, there are two more hydrogens in the
branched (3aa) and linear amide products (4aa) compared to
Scheme 1. Metal-catalyzed aminocarbonylation of olefin with ni-
troarene.

1154
Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
their corresponding substrates. Therefore, hydrogen sources
are one of the key factors in achieving the proposed reductive
aminocarbonylation transformation. Accordingly, we first
screened the influence of the hydrogen source using
K2PdCl4/dppf as the catalytic system (Table 1). Interestingly,
H2O gave an 82% yield of 3aa (entry 1); p-toluene sulfonic acid
showed similar reactivity (entry 2), but other organic acids
showed lower reactivity. Organic boric acid and acetic acid
provided 3aa in 16% (entry 3) and 5% (entry 4) yields, respec-
tively. As observed in Beller’s work, nitroarene was catalytical-
ly reduced to aniline derivatives under H2 atmosphere [35], and
a 68% yield of linear 4aa was obtained as the major amide
product with simultaneous isolation of 17% yield of 3aa (entry
5). A yield of 3aa of only 5% was obtained using the phosphoric
acid as the hydrogen source (entry 6). B(OH)3 was found to be
the best hydrogen source in this catalytic procedure, achieving
a 98% yield of 3aa (entry 7).
Subsequently, with B(OH)3 as the hydrogen source, the in-
fluence of Pd precursors was investigated in the model reaction
(Scheme 2). In general, the halogenated Pd precursors, regard-
less of their neutral or ionic nature, showed good selectivity to
branched amide 3aa, and the ionic K2PdCl4 achieved a near
quantitative yield of the 3aa; on the other hand, other
non-halogenated Pd precursors with different counter anions
displayed different regioselectivity and even chemoselectivity.
The ionic Pd(CH3CN)4X2 with a lower electron affinity anion, for
example, X = BF4 or OTf, provided linear 4aa as the major
product with a yield of approximately 70%. Other normally
used neutral Pd salts containing non-halogen anions (such as
Pd(acac)2 and Pd(NO3)2) revealed completely different
chemoselectivity; only the reduction and reductive carbonyla-
tion of the 3-chloronitrobenzene were observed, yielding
3-chloroaniline (5a) and/or 1-chloro-3-isocyanato-benzene
(6a) as the major products. These investigations indicate that
K2PdCl4 and Pd(CH3CN)4(OTf)2 are excellent Pd precursors for
the synthesis of branched and linear amides, respectively.
As previously mentioned, the ligand has been proved to be a
crucial factor influencing the regioselectivity and activity of the
olefin carbonylation [28–34,39–47]. In the model of aminocar-
bonylation under the optimal K2PdCl4/B(OH)3 catalytic system
shown in Scheme 3 (left part), it was found, surprisingly, that
Scheme 2. Effect of Pd precursors in the reductive aminocarbonylation
of olefin with nitroarene.a aReaction conditions: 1a (0.2 mmol), 2a (0.2
mmol), [Pd] (0.01 mmol), dppf (0.01 mmol), B(OH)3 (0.4 mmol), THF
(2.5 mL), CO (3.5 MPa); 80 °C, 20 h, isolated yield. b5a is 3-chloroaniline.
c6a is 1-chloro-3-isocyanato-benzene.
the phosphine ligands could not affect the aminocarbonylation
regioselectivity but significantly influenced the catalytic activi-
ty; dppf gave 98% yield and 100% selectivity of 3aa. On the
contrary, these phosphine ligands had a significant influence on
the regioselectivity in the corresponding aminocarbonylation
of olefin with amine compound as the N-source; the dppf ligand
yielded linear amide as the major product, and PPh3 had the
best 3aa yield (right part of Scheme 3). These control results
primarily indicate that the present aminocarbonylation of ole-
fin with nitroarene as N-source under the K2PdCl4/B(OH)3 cat-
alytic system does not produce the reduced amine intermedi-
ate, and its regioselectivity control is different from that of the
traditional aminocarbonylation with amine as N-source.
Ionic Pd(CH3CN)4(OTf)2 with an electrophilic –OTf anion
mainly provided linear products (4aa) (Scheme 2), and B(OH)3
was found to be the optimal hydrogen source for the synthesis
of linear 4aa (Table S1). With Pd(CH3CN)4(OTf)2/B(OH)3 as the
catalyst system, the ligand effect on the synthesis of linear
products was investigated, as illustrated in Scheme 4. The lig-
and showed a different regioselectivity control with K2PdCl4 as
the metal precursor. Bidentate phosphines with rigid skeletons
yielded linear amide as the major product; Xantphos was the
optimal ligand, achieving a 92% amide yield and an 80% regi-
oselectivity to 4aa, while monodentate phosphine ligands
tended to form branched amides with low yields. Thus,
Pd(CH3CN)4(OTf)2/Xantphos/B(OH)3 was selected as the cata-
lyst system for the reductive aminocarbonylation of olefin to
linear amide.
Table 1
Screen of hydrogen sources under the catalytic system of K2PdCl4/dppf.
Entry
Hydrogen sources (equiv.)
H2O (2)
Yield
82% 3aa
85% 3aa
16% 3aa
5% 3aa
1
2
3
4
5*
6
7
TsOH·H2O (1)
PhB(OH)2 (2)
AcOH (2)
H2 (1.0 MPa)
H3PO4 (2)
Other reaction conditions, including CO pressure, solvent,
temperature, and reaction time, were also systematically test-
ed. With K2PdCl4/dppf/B(OH)3 as the catalytic system, the am-
ide yields decreased sharply when lowering the CO pressure
(Table S2), but no regioselectivity effects were observed. The
pressure for the synthesis of branched amide (Table S2, entry
1) was set to 3.5 MPa. In the preparation of linear amide under
17% 3aa + 68% 4aa
5% 3aa
B(OH)3 (2)
98% 3aa
Reaction conditions: Styrene (1a, 0.2 mmol), 3-chloronitrobenzene (2a,
0.2 mmol), K2PdCl4 (0.01 mmol), dppf (0.01 mmol), hydrogen source,
THF (2.5 mL), CO (3.5 MPa); 80 °C, 20 h, isolated yield. * CO (2.5 MPa).

Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
1155
Scheme 3. Ligand effect in the reductive aminocarbonylation of olefin with nitroarene and aniline. Reaction conditions: 1a (0.2 mmol), 2a (0.2 mmol)
K2PdCl4 (0.01 mmol), ligands (0.01 or 0.02 mmol), B(OH)3 (0.4 mmol), THF (2.5 mL), CO (3.5 MPa); 80 °C, 20 h, isolated yield.
the Pd(CH3CN)4(OTf)2/Xantphos/B(OH)3 catalytic system, the
CO pressure had little effect on the reactivity, and a lower
pressure led to a higher selectivity (Table S2). A 79% yield of
4aa could be obtained at a CO pressure as low as 0.1 MPa (Ta-
ble S2, entry 5). As presented in Table S3, the solvent polarity
significantly affected both the catalytic activity and selectivity
in the synthesis of branched and linear amides. Highly polar
solvents were advantageous for the hydrogenation of ni-
troarene; only a small amount of carbonylated product was
isolated (Table S3, entries 7–9). The moderately polar THF was
the common solvent for the synthesis of these two amides (Ta-
ble S3, entry 3). The product yield decreased sharply at rela-
tively low temperature without change in regioselectivity (Ta-
ble S4), and the optimal yields of linear and branched amides
were obtained at 80 °C. Finally, the kinetic profile of the reduc-
increased to ~47% before starting to decrease after 5 h, imply-
ing that 3-chloroaniline (5a) may be one intermediate of the
production of linear amide. In conclusion, the optimized condi-
tion for the synthesis of 3aa was condition A: 5 mol% K2PdCl4,
5 mol% dppf, 2 equiv. B(OH)3, 2.5 mL THF, 3.5 MPa CO, 80 °C,
20 h, and condition B: 5 mol% Pd(CH3CN)4(OTf)2, 5 mol%
Xantphos, 2 equiv. B(OH)3, 2.5 mL THF, 0.1 MPa CO, 80 °C, 20 h,
were for the formation of 4aa.
3.2. Substrate scope
For the preparation of the branched amides under condition
A, the scope and limitations of this catalytic system toward
various olefins were first investigated, and the results are de-
picted in Scheme 5. Styrenes with a series of functional groups
tive
aminocarbonylations
of
styrene
(1a)
and
H
3-chloronitrobenzene (2a) were conducted under Condition A
(Fig. S1) and Condition B (Fig. S2), respectively. The results
showed that 20 h was the optimal reaction time for the two
transformations, and compound 3-chloroaniline (5a) was gen-
erated at the beginning under Condition B and then gradually
O
R
N
Cl
O
Cl
NO2
Condition A
CO
+
+
R
R
N
Cl
H
1
2a
3
4
H
H
H
H
N
O
N
Cl
O
N
Cl
O
N
Cl
O
Cl
R1
Me
Cl
R2
Cl
Me
Me
R2 = Me, 3ea, 85%
F, 3fa, 79%
R1 = H, 3aa, 98%
Me, 3ba, 90%
3ca : 4ca = 87 : 13 (85%)
3da : 4da = 87 : 13 (80%)
H
H
H
O
N
Cl
O
N
Cl
O
O
N
Cl
NH
S
Cl
R3
3na: 70%
3oa, 72%
3ma, 80%
H
N
R3 = Me, 3ga, 82%
tBu, 3ha, 92%
OMe, 3ia, 41%
F, 3ja, 89%
O
Cl
H
O
O
N
Cl
O
N
H
Cl
nBu
NC
O
3pa, 92%
3qa, 90%
3ra, 84%
Cl, 3ka, 85%
O
O
H
N
H
Cl
N
Cl
N
Cl
Cl
H
O
O
3ua : 4ua = 16 : 84 (95%)
3ta : 4ta = 12 : 88 (98%)
3sa : 4sa = 15 : 85 (96%)
O
H
H
N
Cl
H
N
N
AcO
Cl
EtO
Cl
4
O
O
O
3xa : 4xa = 0 : 100 (80%)
3va : 4va = 0 : 100 (88%)
3wa : 4wa = 12 : 88 (85%)
H
N
Cl
5
Scheme 4. Ligand effect in the reductive aminocarbonylation of olefin
to linear products. Reaction conditions: 1a (0.2 mmol), 2a (0.2 mmol),
Pd(CH3CN)4(OTf)2 (0.01 mmol), ligands (0.01 or 0.02 mmol), B(OH)3
(0.4 mmol), THF (2.5 mL), CO (0.1 MPa); 80 °C, 20 h, isolated yield.
O
3ya : 4ya = 15 : 85 (98%)
Scheme 5. Olefin scope in the synthesis of branched amides under
Condition A.

1156
Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
on the ortho-, meta-, and para-positions were compatible with
the optimal condition to afford the corresponding branched
products (3aa–3ka) in good to excellent yields and high regi-
oselectivities. Specifically, both electron-rich and elec-
tron-deficient ortho-disubstituted styrenes could also produce
branched products with good yields and high regioselectivity
(3ca and 3da). Interestingly, an increasing steric hindrance of
the vinyl group had little effect on the formation of branched
products with 100% selectivity (3oa), and, notably, no isomer-
ized olefin (allylbenzene) or its aminocarbonylation product
was obtained. Sulfur-containing heterocyclic olefin could also
give a 70% yield of branched amide with 100% selectivity
(3na). Furthermore, excellent yields of 3pa and 3qa were ob-
tained in the reaction of aliphatic alkenes with functional
groups. Representative cyclic alkene demonstrated high activi-
ty, achieving good yield of 3ra. Unexpectedly, when normal
aliphatic olefins (1s–1y) were used, the main products under
Condition A were linear amides, which followed the an-
ti-Markovnikov rules [48,49].
Subsequently, we investigated the scope of olefins under
Condition B, as shown in Scheme 6. The ortho-steric hindrance
of aromatic olefins could increase the regioselectivity for the
synthesis of linear amides under Condition B (4ba, 4ca, and
4da). Good yields of linear products could be achieved by al-
tering the reaction temperature and the CO pressure. Various
functional groups, such as halogens, alkyl, and methoxy, con-
tained in phenyl rings, were well tolerated under Condition B.
The corresponding linear products were achieved with good to
excellent yields and high selectivity. For the more challenging
aliphatic olefins, good to excellent results were also obtained;
moreover, aliphatic olefins with various substituents, such as
ketone, esters, and halogens, were compatible with the catalytic
system under Condition B (4qa, 4ua, 4va, and 4xa). Notably,
this transformation could be applied in the gram-scale synthe-
sis of propanil (propanil is an acetanilide post-emergence
herbicide with no residual effect, and is used against numerous
grasses and broad-leaved weeds [50]), achieving excellent yield
(86%) with a much lower amount of catalyst (0.25 mol%) and
lower CO pressure (0.1 MPa).
Next, we turned our attention to study the scopes of ni-
troarenes using styrene (1a) as a standard coupling partner
(Scheme 7). In previous reports, the aminocarbonylation was
sensitive to the steric hindrance and electronic effects of the
substituents on the anilines [35], but there were no obvious
effects on the aminocarbonylation when using nitroarenes as
nitrogen sources under Condition A. For example, the elec-
tron-rich or electron-deficient ortho-substituted derivatives
gave good yields with 100% branched regioselectivity
(3ad–3af). The steric hindrance of nitroarenes slightly reduced
the activity under Condition B with excellent selectivity for the
linear products (4ad–4af). The nitroarenes with nitrile (3ag),
ester (3ah), and ketone (3ai) moieties worked efficiently under
Condition A. Notably, with this procedure, which is perfectly
suitable for the substrates with an aldehyde substituent,
providing corresponding branched amide (3aj) with 90% yield
and pure branched regioselectivity, it was usually necessary to
protect the aldehyde group when the amine was used as a ni-
trogen source in the aminocarbonylation, particularly under
the acid conditions. Other typical functional groups on the ni-
troarenes, such as halogen, ether, and alkyl, were also well tol-
erated under Condition A, producing the corresponding
branched products in 88%–95% yields with 100% branched
regioselectivity (3ak–3am). Aminocarbonylation with nitro-
H
O
O
R
N
Cl
Cl
NO2
Condition B
CO
+
+
R
R
N
Cl
H
1
2a
4
3
R1
O
Me
O
Cl
O
N
Cl
N
Cl
N
Cl
H
H
H
Cl
Me
Me
R1 = H, 4aa : 3aa = 80 : 20 (98%)
Me, 4ba : 3ba = 98 : 2 ( 92%)
4ca, 80%a
4da : 3da = 87 : 13 (75%)a
O
O
O
N
R2
N
H
Cl
N
Cl
Cl
Cl
H
H
S
4ma, 60%b
R2 = Me, 4ea, 88%
4na, 40%b
F, 4fa, 82%
O
O
O
O
nBu
N
H
N
H
Cl
N
H
Cl
O
R3
R3 = Me, 4ga, 85%
4oa : 0a
4qa : 3qa = 90 : 10 (94%)
tBu, 4ha : 3ha = 89 : 11 (92%)
OMe, 4ia : 3ia = 92 : 8 (53%)
F, 4ja, 92%
H
O
N
Cl
Ph
N
Cl
O
H
Cl, 4ka : 3ka = 85 : 15 (98%)
CF3, 4la, 95%
4ra, 94%
4sa, 86 %
O
N
O
O
Ac
Cl
Ph
N
Cl
3
3
AcO
N
Cl
4
H
H
H
4va : 3va = 80 : 20 (98%)
4ua, 86 %
4ta, 72 %
O
O
H
N
O
Cl
Cl
EtO
Cl
Cl
N
Cl
N
H
Cl
O
N
6
3
5
H
H
O
4wa : 3wa = 92 : 8 (98%)
Propanil, 86 %c
4ya : 3ya = 90 : 10 (97%)
4xa : 3xa = 90 : 10 (98%)
Scheme 6. Olefin scope in the synthesis of linear amides underCondi-
tion B. aCO (3.5 MPa); 100 °C. bCO (3.5 MPa). c1 (1.5 MPa), 2 (10 mmol),
Pd(CH3CN)4(OTf)2 (0.025 mmol), Xantphos (0.025 mmol), B(OH)3 (20
mmol), THF (25 mL), CO (0.1 MPa); 80 °C, 20 h.
Scheme 7. Nitroarene scope in the synthesis of branched and linear
amides under Condition Aa and Condition Bb.

Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
1157
thiophene also proceeded smoothly with a 70% yield of the
desired 3an and 100% regioselectivity (3an). Under Condition
B, the universality of nitro compounds was not as good as un-
der Condition A. Electron-donating groups on the nitro com-
pound slightly lowered the amide yields (4ab, 4af and 4am),
and the nitrobenzene with a 4-aldehyde moiety (4aj) and ni-
trothiophene did not work under Condition B.
were realized using aromatic amine and isocyanate as
N-source, unlike in the aminocarbonylation of nitroarene
(Scheme 8(a)iv and 8(a)vi), but similar results were obtained
using these three N-precursors under Condition B (Scheme
8(b)i, 8(b)iii, and 8(b)v). Thus, it is deduced that the production
of linear amide possibly involves aromatic amine and/or aryl
isocyanate, which, however, are not the intermediates in the
synthesis of branched amide. This could be further verified
from the previous result where H2 was used as a hydrogen
source (Table 1, entry 4). Furthermore, the generation of
3-chloroaniline (5a) and its further carbonylation to linear 4aa
were observed in the kinetic profile of the model reaction un-
der Condition B (Fig. S2), indicating that the amine may be the
intermediate for the production of the linear product. Very low
yields of amides were isolated when N-aryl hydroxylamine was
used under Condition A and Condition B (Scheme 8(a)v and
8(b)iv), possibly suggesting that N-aryl hydroxylamine was not
the intermediate in the present carbonylations.
In the reductions of nitroarenes under CO atmosphere, pal-
ladium imido species “Pd=NPh” was usually postulated as the
key intermediate [54,55]. To seek the evidence for the exist-
ence of “Pd=NPh,” attempts were made to trap this species by
adding cyclohexene dropwise during a catalytic experiment
under Condition A (Fig. 1) and Condition B (Scheme S1)
[17,56]. As indicated by the mass and fragmentation pattern of
the peak in the GC-MS spectrum (Fig. 1), the corresponding
aziridine (7-(3-chlorophenyl)-7-azabicyclo[4.1.0]heptane) was
indeed observed under Condition A. However, this trapping
was not detected under Condition B (Scheme S1). Therefore, it
is believed that the palladium imido species was the key inter-
mediate for the formation of branched amides. It has been re-
ported that azobenzenes could react with lower-valent metals
to obtain the corresponding metal imido species [7,9,57,58]. To
3.3. Mechanistic insights
The above differences in regioselectivity, reaction parame-
ter effect, and substrate scope seem to indicate that the present
selective carbonylations undergo different reaction pathways
under Condition A and Condition B. It has been found that ni-
troarene can be catalytically reduced to nitrosoarene, aromatic
amine, N-aryl hydroxylamine, and aryl isocyanate under CO
atmosphere [16,17,51–53]. Therefore, control reactions with
these N-species would provide more insights into the present
aminocarbonylation with nitroarene as N-source. As illustrated
in Scheme 8, when replacing nitroarene with its first reduced
intermediate (nitrosoarene) under Condition A, the branched
amide was obtained with a yield as small as 27% (Scheme
8(a)ii) while linear amide was obtained with a 68% yield. The
yield of branched products could be significantly enhanced to
87% by dropwise addition of nitrosoarene over 15 h (Scheme
8(a)iii). This yield is close to that obtained under Condition A
(Scheme 8(a)i), wherein the nitrosoarene was supposed to
slowly release from the catalytic reduction of nitroarene, indi-
cating that, under Condition A, the conversion of nitrosoarene
to aniline was faster than its reductive aminocarbonylation. In
the meantime, similar catalytic results were obtained in the
aminocarbonylation of nitrosoarene and nitroarene under
Condition B (Scheme 8(b)i and 8(b)ii)). In light of the results of
this and previous works, it is proposed that nitrosoarene is the
common N-intermediate for the synthesis of branched and
linear amides. Under Condition A, similar product distributions
further
support
this
hypothesis,
Cl
Cl
N
PdP2
Condition A
N
+
(a) Aminocarbonylation of styrene with different N-sources under Condition A
GC-MS detected
H
O
O
N
Cl
NH2
Ar
Ar
Condition A
CO
Ar [N]
+
+
Ph
N
Ph
H
Ph
(i)
Ar-NO2
Ar-NO
98%
0
(ii)
27%
87%
20%
17%
16%
68%
0
Cl
N
(iii)a
(iv)
(v)
Ar-NO
Ar-NH2
Ar-NHOH
Ar-NCO
80%
0
(vi)
52%
Cl
N
(b) Aminocarbonylation of styrene with different N-sources under Condition B
H
O
O
N
Ar
Condition B
CO
Ar
Ph
(i)
(ii)
(iii)
(iv)
(v)
Ar [N]
+
+
Ph
N
H
Ph
Ar-NO2
Ar-NO
Ar-NH2
Ar-NHOH
Ar-NCO
74%
66%
56%
14%
52%
18%
18%
14%
7%
Cl
N
8%
Scheme 8. Studies on the possible N-intermediates involved in the
reductive aminocarbonylation. Ar = 3- chlorophenyl. aReaction in 5
mmol; 1-chloro-3-nitrosobenzene was added dropwise over 15 h.
m / z
Fig. 1. The trapping experiment of an imido complex with cyclohexene.

1158
Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
1,2-bis(3-chlorophenyl)diazene was used as the substrate to
conduct the control experiments. A 62% yield of branched 3aa
was isolated under Condition A, and no reaction occurred un-
der Condition B (Scheme S2). These results show that different
N-intermediates were involved in the production of branched
and linear products. Nitrene and its coordinated palladium
imido species are the possible intermediates for the regioselec-
tive synthesis of branched amide.
Scheme 9. Control experiment to further verify the production of 7aa
intermediate.
In the ligand optimization for the preparation of branched
products, N-(3-chlorophenyl)-2-phenylacrylamide (7aa) was
detected by GC-MS when tris(4-methoxyphenyl) phosphine
and DPEPhos were used as ligands. To probe whether 7aa was
an intermediate or a byproduct, the reaction rate was reduced
by lowering the CO pressure, as indicated in the model reaction
(Fig. 2), Formation of 7aa was observed at the beginning; it
gradually increased to 12% and kept this concentration before
starting to decrease after 20 h, implying that 7aa was one pos-
sible intermediate for the production of branched amide. Sub-
sequently, 7aa was used as the substrate under Condition A,
and 3aa was produced with a 92% yield (Scheme 9). These
results suggest that enamide was an intermediate for the syn-
thesis of branched amide.
Based on our investigations and previous works
[5–15,37,38], the following catalytic cycles for the synthesis of
branched and linear products were proposed, as shown in
Scheme 10. The reductive aminocarbonylation was initiated by
the reduction of nitroarene with CO via Pd-nitrosoarene (II)
complex under both condition A and condition B [17,38]. Com-
plete deoxygenation of the Pd-nitrosoarene (II) complex yield-
ed nitrene intermediate (III) with CO2 releasing under Condi-
tion A [17,38], which was verified by the situ-trapping experi-
ment (Fig. 1). Subsequently, the addition of olefins to nitrene
intermediate (III) and the insertion of CO occurred, followed by
the reductive elimination to yield an enamide, which was re-
ductively catalyzed by the eliminated Pd(0) and CO with the
help of B(OH)3 to provide the branched product 3. This step
could be reasonably deduced by the reaction profile (Fig. 2)
and the controlled experiment (Scheme 9).
For the synthesis of linear products, the hydrodeoxygena-
tion of the Pd-nitrosoarene (II) complex yielded the aromatic
amine intermediate (5) [16,17], which was verified by the reac-
tion profile for the synthesis of linear products (Fig. S2) and the
controlled experiment (Scheme 8(b)iii). Subsequently, it fol-
lowed a similar route in the reported aminocarbonylation of
olefin with amine to produce linear amide (4) [30].
4. Conclusions
In summary, we have developed a route for the Pd-catalyzed
selective aminocarbonylation of olefins with nitroarenes via
different
intermediates.
Both
the
chemoselectivity
(3-chloroaniline/1-chloro-3-isocyanato-benzene or amide) and
the carbonylative regioselectivity are sensitive to the nature of
counter anions of Pd(II) precursors. The nitroarene was com-
pletely deoxidized to nitrene and regioselectively carbonylated
with olefin to yield branched amide using K2PdCl4 as the Pd
precatalyst. The phosphine ligand hardly affected the car-
bonylation regioselectivity but affected the reactivity. On the
other hand, the kinetic profile and controlled experiments re-
vealed that nitroarene is first hydrogenated to aromatic amine,
which is the essential intermediate for the production of linear
products
under
the
optimal
Pd(CH3CN)4(OTf)2/
Xantphos/B(OH)3 catalytic system. These catalytic systems
provide facile and selective access to linear and branched am-
ides and will aid the development of diverse synthetic routes
through the formation of different N-intermediates.
2a
3aa
7aa
110
100
90
80
70
60
50
40
30
20
10
0
-10
0
5
10
15
20
25
30
35
Time / h
Scheme 10. Proposed reaction pathways for the selective aminocar-
bonylation.
Fig. 2. Reaction profile for the synthesis of 3aa.

Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
1159
8943–9021.
References
[24] J. B. Peng, H. Q. Geng, F. P. Wu, D. Li, X.-F. Wu, J. Catal., 2019, 375,
519–523.
[25] J. W. Li, S. Y. Wang, S. C. Zou, H. M. Huang, Commun. Chem., 2019, 2,
1–9.
[26] B. Gao, G. Y. Zhang, X. B. Zhou, H. M. Huang, Chem. Sci., 2018, 9,
380–386.
[27] L. Yang, L. J. Shi, Q. Xing, K. W. Huang, C. G. Xia, F. W. Li, ACS Catal.,
2018, 8, 10340–10348.
[28] F. Sha, H. Alper, ACS Catal., 2017, 7, 2220–2229.
[29] J. Liu, H. Q. Li, A. Spannenberg, R. Franke, R. Jackstell, M. Beller,
Angew. Chem. Int. Ed., 2016, 55, 13544–13548.
[30] T. Y. Xu, F. Sha, H. Alper, J. Am. Chem. Soc., 2016, 138,
6629–6635.
[31] X. Y. Li, X. W. Li, N. Jiao, J. Am. Chem. Soc., 2015, 137, 9246–9249.
[32] H. Q. Li, K. W. Dong, H. Neumann, M. Beller, Angew. Chem. Int. Ed.,
2015, 54, 10239–10243.
[1] J. S. Zhu, C. J. Li, K. Y. Tsui, N. Kraemer, J.-H. Son, M. J. Haddadin, D. J.
Tantillo, M. J. Kurth, J. Am. Chem. Soc., 2019, 141, 6247–6253.
[2] M. Rauser, R. Eckert, M. Gerbershagen, M. Niggemann, Angew.
Chem. Int. Ed., 2019, 58, 6713–6717.
[3] J. H. Gui, C. M. Pan, Y. Jin, T. Qin, J. C. Lo, B. J. Lee, S. H. Spergel, M. E.
Mertzman, W. J. Pitts, T. E. L. Cruz, M. A. Schmidt, N. Darvatkar, S.
R. Natarajan, P. S. Baran, Science, 2015, 348, 886–891.
[4] N. Ono, The Nitro Group in Organic Synthesis, Wiley-VCH, Wein-
heim, 2001, 1–2.
[5] C. W. Cheung, J. A. Ma, X. L. Hu, J. Am. Chem. Soc., 2018, 140,
6789–6792.
[6] T. V. Nykaza, A. Ramirez, T. S. Harrison, M. R. Luzung, A. T. Ra-
dosevich, J. Am. Chem. Soc., 2018, 140, 3103–3113.
[7] C. W. Cheung, M. L. Ploeger, X. L. Hu, Nat. Commun., 2017, 8,
14878–14887.
[33] G. Y. Zhang, B. Gao, H. M. Huang, Angew. Chem. Int. Ed., 2015, 54,
7657–7661.
[34] K. W. Dong, X. J. Fang, R. Jackstell, G. Laurenczy, Y. H. Li, M. Beller, J.
Am. Chem. Soc., 2015, 137, 6053–6058.
[35] X. J. Fang, R. Jackstell, M. Beller, Angew. Chem., Int. Ed., 2013, 52,
14089–14093.
[36] J. B. Peng, H. Q. Geng, D. Li, X. X. Qi, J. Ying, X. F. Wu, Org. Lett.,
2018, 20, 4988–4993.
[37] F. Zhou, D. S. Wang, X. Y. Guan, T. G. Driver, Angew. Chem. Int. Ed.,
2017, 56, 4530–4534.
[38] C. W. Cheung, M. Leendert Ploeger, X. L. Hu, Chem. Sci., 2018, 9,
655–659.
[8] M. Shevlin, X. Y. Guan, T. G. Driver, ACS Catal., 2017, 7, 5518–5522.
[9] C. W. Cheung, M. L. Ploeger, X. L. Hu, ACS Catal., 2017, 7,
7092–7096.
[10] R. Rubio-Presa, M. R. Pedrosa, M. A. Fernandez-Rodríguez, F. J.
Arnaiz, R. Sanz, Org. Lett., 2017, 19, 5470–5473.
[11] Z. C. Lin, Z. Y. Hu, X. Zhang, J. H. Dong, J. B. Liu, D. Z. Chen, X. X. Xu,
Org. Lett., 2017, 19, 5284–5287.
[12] M. Rauser, C. Ascheberg, M. Niggemann, Angew. Chem. Int. Ed.,
2017, 56, 11570–11574.
[13] C. W. Cheung, X. L. Hu, Nat. Commun., 2016, 7, 12494.
[14] K. L. Zhu, M. P. Shaver, S. P. Thomas, Chem. Sci., 2016, 7,
3031–3035.
[39] H. R. Liang, L. Zhang, X. L. Zheng, H. Y. Fu, M. L. Yuan, R. X. Li, H.
Chen, Chin. J. Catal., 2012, 33, 977–981.
[15] N. Jana, F. Zhou, T. G. Driver, J. Am. Chem. Soc., 2015, 137,
6738–6741.
[40] A. C. Brezny, C. R. Landis, J. Am. Chem. Soc., 2017, 139, 2778–2785.
[41] W. L. Ren, W. J. Chang, J. Dai, Y. Shi, J. F. Li, Y. A. Shi, J. Am. Chem.
Soc., 2016, 138, 14864–14867.
[42] D. Semeril, C. Jeunesse, D. Matt, L. Toupet, Angew. Chem. Int. Ed.,
2006, 45, 5810–5814.
[43] S. C. Bourque, F. Maltais, W. J. Xiao, O. Tardif, H. Alper, P. Arya, L. E.
Manzer, J. Am. Chem. Soc., 1999, 121, 3035–3038.
[44] J. B. Peng, X. F. Wu, Angew. Chem. Int. Ed., 2018, 57, 1152–1160.
[45] D. Ding, G. H. Zhu, X. F. Jiang, Angew. Chem. Int. Ed., 2018, 57,
9028–9032.
[16] T. J. Mooibroek, E. Bouwman, E. Drent, Organometallics, 2012, 31,
4142–4156.
[17] T. J. Mooibroek, L. Schoon, E. Bouwman, E. Drent, Chem. Eur. J.,
2011, 17, 13318–13333.
[18] K.-S. Ju, R. E. Parales, Microbiol. Mol. Biol. Rev., 2010, 74, 250–272.
[19] A. M. Tafesh, J. Weiguny, Chem. Rev., 1996, 96, 2035–2052.
[20] R. M. Figueiredo, J. S. Suppo, J. M. Campagne, Chem. Rev., 2016,
116, 12029–12122.
[21] V. R. Pattabiraman, J. W. Bode, Nature, 2011, 480, 471–479.
[22] Q. J. Yuan, M. S. Sigman, J. Am. Chem. Soc., 2018, 140, 6527–6530.
[23] X. G. Guo, A. Facchetti, T. J. Marks, Chem. Rev., 2014, 114,
[46] D. Ding, T. Mou, M. H. Feng, X. F. Jiang, J. Am. Chem. Soc., 2016, 138,
Graphical Abstract
Chin. J. Catal., 2020, 41: 1152–1160 doi: 10.1016/S1872-2067(20)63561-6
Intermediate formation enabled regioselective access to
amide in the Pd-catalyzed reductive aminocarbonylation
of olefin with nitroarene
Li Yang, Lijun Shi, Chungu Xia *, Fuwei Li *
Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of
Sciences
An efficient palladium-catalyzed reductive aminocarbonylation of
olefins with nitroarenes was developed, providing facile access to
amides with excellent regioselectivities enabled by the different
N-intermediate formation reduced from nitroarene. This route is
significantly different from the previous ligand-controlled regiose-
lective synthesis methods of amides via aminocarbonylation of
olefins with amines.

1160
Li Yang et al. / Chinese Journal of Catalysis 41 (2020) 1152–1160
5218–5221.
[53] F. Ragaini, M. Gasperini, S. Cenini, L. Arnera, A. Caselli, P. Macchi, N.
Casati, Chem. Eur. J., 2009, 15, 8064–8077.
[54] F. Paul, Coord. Chem. Rev., 2000, 203, 269–323.
[55] R. Waterman, G. L. Hillhouse, J. Am. Chem. Soc., 2003, 125,
13350–13351.
[47] M. H. Feng, B. Q. Tang, N. Z. Wang, H. X. Xu, X. F. Jiang, Angew.
Chem. Int. Ed., 2015, 54, 14960–14964.
[48] H. Q. Li, K. W. Dong, H. J. Jiao, H. Neumann, R. Jackstell, M. Beller,
Nat. Chem., 2016, 8, 1159–1166.
[49] M. Beller, J. Seayad, A. Tillack, H. J. Jiao, Angew. Chem. Int. Ed.,
2004, 43, 3368–3398.
[50] S. Matsunaka, Science, 1968, 160, 1360–1361.
[51] A. Vavasori, L. Ronchin, Pure Appl. Chem., 2012, 84, 473–484.
[52] Y. Takebayashi, K. Suea, S. Yodaa, T. Furuyaa, K. Mae, Chem. Eng. J.,
2012, 180, 250–254.
[56] S. M. Bellows, N. A. Arnet, P. M. Gurubasavaraj, W. W. Brennessel,
E. Bill, T. R. Cundari, P. L. Holland, J. Am. Chem. Soc., 2016, 138,
12112–12123.
[57] M. A. Aubart, R. G. Bergman, Organometallics, 1999, 18, 811–813.
[58] C. Amatore, E. Carre, A. Jutand, M. A. M'Barki, Organometallics,
1995, 14, 1818–1826.
钯催化芳香硝基化合物与烯烃的选择性还原羰化酰胺化反应
杨 莉, 石利军, 夏春谷*, 李福伟#
中国科学院兰州化学物理研究所, 羰基合成与选择氧化国家重点实验室, 甘肃兰州730000
摘要: 与芳香胺相比, 芳香硝基化合物具有廉价易得、官能团兼容性好等优点, 作为氮源在下游含氮化学品合成中具有广
泛的应用. 目前烯烃羰化酰胺化反应绝大多数以胺类化合物为氮源, 其中直链和支链酰胺产物的选择性主要是通过具有
特定电子和位阻特性的配体调控实现. 已报道的芳香硝基化合物的还原酰胺化反应研究中, 需要外加还原剂或者利用金
属羰基化合物Mo(CO)6释放的CO为羰基源和还原剂. 本文发展了一种毋须外加还原剂的钯催化芳香硝基化合物与烯烃的
还原羰化酰胺化反应新方法.
研究发现, 钯金属催化剂(特别是离子型)的抗衡阴离子是还原羰化酰胺化反应中化学选择性和羰化区域选择性的关键
因素. 抗衡阴离子为氯离子、硼酸为助剂时, 最优钯前驱物K2PdCl4的产物主要为支链酰胺, 此时不同的膦配体并不能调控
其区域选择性, 这与胺的烯烃酰胺化反应可以通过配体调控羰化的区域选择性表现出明显的不同. 含氮中间体原位捕捉、
硝基化合物还原下游可能中间体对照实验等研究表明, 芳香硝基化合物在以一氧化碳为还原剂的催化还原体系下被完全
脱氧还原为氮烯(Ar-N:), 再经过烯酰胺中间体进一步烯键还原得到相应的支链酰胺; 当离子型钯前驱物的抗衡阴离子配
位性较弱时, 最优钯前驱物为Pd(CH3CN)4(OTf)2时, 以直链酰胺为主要产物, 此时不同的膦配体可以调控酰胺化的区域选
择性. 同样的机理研究表明, 在该催化剂体系下芳香硝基化合物首先被还原为芳基胺, 然后再发生与现有报道类似的胺类
化合物的烯烃羰化酰胺化反应. 这两个催化反应体系都表现出了较好的底物适用性, 并且可以高效地应用于除草剂(敌稗)
的一步合成. 本文为以硝基化合物为起始氮源, 通过催化控制生成特定含氮中间体, 从而可控合成不同的含氮化学品提供
了一条新思路.
关键词: 芳香硝基化合物; 酰胺; N-中间体; 氨基羰基化; 区域选择性
收稿日期: 2019-10-22. 接受日期: 2019-12-01. 出版日期: 2020-07-05.
*通讯联系人. 电子信箱: cgxia@licp.cas.cn
#通讯联系人. 电子信箱: fuweili@licp.cas.cn
基金来源: 国家重点研发计划(2018YFB1501600); 国家自然科学基金(21773271, 21972151).
本文的电子版全文由Elsevier出版社在ScienceDirect上出版(http://www.sciencedirect.com/science/journal/18722067).