Aminocarbonylation of aryl halides using a nickel phosphite catalytic system

Article abstract of DOI:10.1021/ol702058e

(Formula Presented) The nickel and phosphite catalytic system with sodium methoxide enables a very efficient aminocarbonylation reaction to be performed between aryl iodides or bromides and N,N-dimethylformamide (DMF). Phosphite ligand 1, which is very st

Full text of DOI:10.1021/ol702058e

ORGANIC
LETTERS
2
007
Vol. 9, No. 22
615-4618
Aminocarbonylation of Aryl Halides
Using a Nickel Phosphite Catalytic
System
4
†
†
†
‡
,†
Jinhun Ju, Miso Jeong, Jeongju Moon, Hyun Min Jung, and Sunwoo Lee*
Department of Chemistry, Chonnam National UniVersity, 300 Yongbong-dong,
Buk-gu, Gwangju 500-757, Republic of Korea, and AdVanced Materials DiVision,
Korea Research Institute of Chemical Technology, 100 Jang-dong, Yuseong-gu,
Daejeon 305-600, Republic of Korea
sunwoo@chonnam.ac.kr
Received August 30, 2007
ABSTRACT
The nickel and phosphite catalytic system with sodium methoxide enables a very efficient aminocarbonylation reaction to be performed
between aryl iodides or bromides and N,N-dimethylformamide (DMF). Phosphite ligand 1, which is very stable to air and moisture and, furthermore,
inexpensive, afforded the highest reaction yield.
Heck reported the first formation of palladium-catalyzed
amides from aryl halides by reaction with carbon monoxide
and primary or secondary amines. Carbon monoxide is
Only a few results have been reported for the use of DMF
as a source of carbon monoxide or amide.
8
1
Therefore, we focused on optimizing the process condi-
tions to give a more practical and inexpensive catalytic
system. Recently, we reported that phosphites can be used
as ligands in palladium-catalyzed Hiyama cross-coupling
generally employed in this reaction as a source of carbonyl
2
groups. To eliminate the cumbersome handling of toxic
3
carbon monoxide gas, a variety of surrogates have been
4
5
6
9
used, including Ni(CO)
4
, Mo(CO)
6
, carbamoylstannanes,
reactions. In this paper, we report the first nickel-catalyzed
7
8
carbamoylsilanes, and dimethylformamide (DMF).
aminocarbonylation of aryl halides using DMF as an amide
source and phosphite as a ligand (Figure 1). To the best of
our knowledge, nickel has never been reported as a catalyst
However, each substitute has its own drawbacks, including
6
thermal instability, reagent cost, the limited scope of aryl
1
0
8
b
8a
in this transformation.
halides, the necessity of using a microwave, or the fact
that the ligand alkylphosphine8 is unstable in air. In
addition, all of these choices rely on palladium as a catalyst.
a,b
(5) (a) Kaiser, N. F. K.; Hallberg, A.; Larhed, M. J. Comb. Chem. 2002,
4, 109-111. (b) Georgsson, J.; Hallberg, A.; Larhed, M. J. Comb. Chem.
2
003, 5, 350-352. (c) Wannberg, J.; Larhed, M. J. Org. Chem. 2003, 68,
5
750-5753. (d) Wu, X.; Ekegren, J. K.; Larhed, M. Organometallics 2006,
†
Chonnam National University.
Korea Research Institute of Chemical Technology.
25, 1434-1439. (e) Wu, X.; Wannberg, J.; Larhed, M. Tetrahedron 2006,
62, 4665-4670.
‡
(
1) (a) Schoenberg, A.; Bartoletti, I.; Heck, R. F. J. Org. Chem. 1974,
(6) Lindsay, C. M.; Widdowson, D. A. J. Chem. Soc., Perkin Trans. 1
1988, 569-573.
3
3
9, 3318-3326. (b) Schoenberg, A.; Heck, R. F. J. Org. Chem. 1974, 39,
327-3331.
(7) (a) Cunico, R. F.; Maity, B. C. Org. Lett. 2002, 4, 4357-4359. (b)
Cunico, R. F.; Maity, B. C. Org. Lett. 2003, 5, 4947-4949. (c) Cunico, R.
F.; Pandey, R. K. J. Org. Chem. 2005, 70, 9048-9050.
(8) (a) Wan, Y.; Alterman, M.; Larhed, M.; Hallberg, A. J. Org. Chem.
2002, 67, 6232-6235. (b) Hosoi, K.; Nozaki, K.; Hiyama, T. Org. Lett.
2002, 4, 2849-2851.
(2) (a) Beller, M.; Cornils, B.; Frohning, C. D.; Kohlpaintner, C. W. J.
Mol. Catal. A.: Chem. 1995, 104, 17-85. (b) Yamamoto, A.; Kayaki, Y.;
Nagayama, K.; Shimizu, I. Synlett 2000, 925-937.
(
3) Morimoto, T.; Kakiuchi, K. Angew. Chem., Int. Ed. 2004, 43, 5580-
5
1
588.
(
4) Corey, E. J.; Hegedus, L. S. J. Am. Chem. Soc. 1969, 91, 1233-
(9) Ju, J.; Nam, H.; Jung, H. M.; Lee, S. Tetrahedron Lett. 2006, 47,
8673-8678.
234.
1
0.1021/ol702058e CCC: $37.00
© 2007 American Chemical Society
Published on Web 10/04/2007

et al. reported that small substituents favor dimeric nickel
1
0c
species, which may not be catalytically active. Mono and
chelating arylphosphine ligands gave low yields (entries 10-
1
2). Noticeably, the homocoupling of aryl halides was not
detected in the nickel-catalyzed reactions; this side reaction
gave rise to the major products in palladium-catalyzed
reactions which afforded no amide-coupled product (entry
1
3).
We next examined the effects of solvents, bases, temper-
atures, and the molar ratios of nickel and ligand in the model
reaction. The results are summarized in Table 2.
Figure 1. Phosphite ligands.
Table 2. Optimization of Aminocarbonylation with Phosphite
a
1
and Ni(OAc)
2
2
‚4H O
To investigate the activity of nickel as a catalyst, we first
screened four different nickel sources and ligands in the
aminocarbonylation of 4-bromotoluene with DMF in the
yieldb (%)
entry
Ni/L
mol %
base
solvent
1
2
3
4
5
6
7
8
9
0
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:1
1:2
1:0.5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
NaOMe
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
diglyme
DMA
99
56
56
19
0c
presence of base. As shown in Table 1, Ni(OAc)
2 2
‚4H O was
t
NaO Bu
KOMe
t
KO Bu
n-BuLi
Table 1. Screening of Nickel Sources and Ligandsa
c
NaHMDS
Na2CO3
Na3PO4
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
NaOMe
0
c
0
c
0
d
25
60
e
1
b
11
93
77
70
97
8
82
78
18
60
72
entry
Ni catalysts
ligands
yield (%)
12
13
14
15
16
17
18
19
20
1
2
3
4
5
6
7
8
9
0
1
2
3
Ni(OAc)2‚4H2O
NiCl2
Ni(acac)2
1
1
1
1
2
3
99
79
52
40
90
87
89
23
0
12
14
43
0e
NMP
DMF
p-xylene
dioxane
dioxane
dioxane
dioxane
dioxane
Ni(COD)2
2.5
1
0.1
5
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Ni(OAc)2‚4H2O
Pd2(dba)3
t
P Bu3
PCy3
P(OEt)3
PPh3
DPPF
Xantphosd
5
a
Reaction conditions: 1.0 mmol of 4-bromotoluene, 4.0 mmol of base,
.5 mL of DMF, 3 mL of solvent, 5 mol % of 1, and 5 mol % of
1
1
1
1
0
c
b
Ni(OAc)2‚4H2O were reacted at 110 °C for 12 h. Yield was determined
c
by GC by comparison to an internal standard (naphthalene). No reaction:
only starting materials appeared in GC. d 2 equiv of NaOMe was used. 3
e
1
equiv of NaOMe was used.
a
Reaction conditions: 1.0 mmol of aryl halide, 4.0 mmol of NaOMe,
.5 mL of DMF, 3 mL of 1,4-dioxane, 5 mol % of nickel, and 5 mol % of
0
b
ligand were reacted at 110 °C for 12 h. Yield was determined by GC by
c
comparison to an internal standard (naphthalene). 1,1′-Bis(diphenylphos-
The type of base was the most important factor in this
reaction. An excellent yield was obtained with NaOMe as
base (entry 1), while other alkoxide or hydroxide bases such
as NaO Bu, KOMe, and KO Bu showed low yields (entries
-4). However, the non-alkoxide-type, strong bases failed
d
phino)ferrocene. 4,5-Bis(diphenylphosphino)-9,9-dimethylxan-
e
thene. No product and 33% of homocoupling product of 4-bromotoluene.
t
t
2
the best nickel source (entry 1). The use of other nickel
to give the desired product (entries 5 and 6). With weak
inorganic bases, no desired product was isolated (entries 7
and 8). Four equivalents of base was required to reach the
highest yield. When using 2 or 3 equiv of NaOMe, the yield
was 25% and 60%, respectively (entries 9 and 10). We then
turned our attention to the cosolvent used with DMF.
Hiyama’s research group described that Pd-catalyzed ami-
nocarbonylation was carried out in DMF without any other
cosolvents. Dioxane was proven to be the best cosolvent,
while diglyme, possessing an ether group like dioxane, also
showed a high yield (entry 11). The reactions with other
amide-type cosolvents resulted in moderate yields (entries
2 2 2
catalysts, such as NiCl , Ni(acac) , and Ni(COD) , gave very
low reaction yields (entries 2-4). Among the ligands
examined, phosphite 1 gave the highest yield. Sterically
t
hindered phosphites and P Bu
3
afforded good yields of more
afforded a poor
showed no activity, possibly due to the
small substituent on the phosphite (entries 8 and 9). Wang
than 87% (entries 5-7). However, PCy
yield, and P(OEt)
3
3
8
b
(10) For a recently reported nickel/phosphite catalytic system for coupling
reactions, see: (a) Ho, C.-Y.; Jamison, T. F. Angew. Chem., Int. Ed. 2007,
4
6, 782-785. (b) Gavryushin, A.; Kofink, C.; Manolikakes, G.; Knochel,
P. Org. Lett. 2005, 7, 4871-4874. (c) Wang, Y.; Marrocco, M. L., III;
Trimmer, M. S. WO 9639455 A1, 1996.
4616
Org. Lett., Vol. 9, No. 22, 2007

12 and 13). However, p-xylene, which was a good solvent
in the Hiyama reaction with phosphite, was not suitable for
the coupling reactions (entry 15). When the amount of
catalyst used was less than 1 mol %, very low yields of the
desired products were obtained (entry 18). A 1:1 ratio of
catalyst to ligand was proven to form the most active catalytic
system, whereas systems with other nickel to phosphite ratios
such as 1:2 or 1:0.5 produced inefficient catalysis for this
reaction (entries 19 and 20).
Table 3. Nickel-Catalyzed Aminocarbonylation of Aryl
Halidesa
To explore the scope of this reaction system, we screened
the aminocarbonylation of representative aryl halides with
DMF (Table 3). Aryl iodides possessing an electron-donating
group, such as a methoxy or methyl group, gave the desired
coupling product in high yield (entries 1 and 2). 2-Iodo-
toluene, bearing a sterically hindered group, afforded the
coupling product in moderate yield (entry 3). 1-Iodonaph-
thalene and 4-iodobenzotrifluoride possessing an electron-
withdrawing group afforded the coupling product in low yield
and produced dehalogenated product in 38% and 45% yields,
respectively (entries 4 and 5). We conclude that, in general,
aryl bromides possessing an electron-donating group gave a
better yield than those having an electron-withdrawing group.
Among the bromoanisoles, the o-bromoanisole, which is
sterically hindered, showed almost quantitative yield (99%)
(entry 9). However, use of the meta version of this molecule
decreased the yield to 82% and use of the para molecule
reduced the yield to 62% (entries 10 and 11). Bromobiphenyl
substrates showed good to excellent yields (entries 12 and
13). The coupling yield of 1-bromonaphthalene was 64%
(entry 14) and that of 2-bromonaphthalene, a less sterically
hindered variant, was 72% (entry 15). Bromonaphthalene,
having methoxy as an electron-donating group, yielded 97%
(
entry 16). Half (51%) of the 2-bromothiophene was
converted to give the desired product in 50% yield, whereas
the yield was dramatically increased to 94% by running the
reaction using 10 mol % catalyst (entry 17).
To understand the reaction mechanism, comparable reac-
tions were carried out on the basis of reaction mechanisms
suggested by palladium-catalyzed aminocarbonylation. The
influence of the electronic properties of the aryl group and
of the base used in the catalytic reactions were investigated.
First, based on the mechanism proposed by Halberg of
8a
adding imidazole and benzylamine as external amine sources,
we obtained neither the imidazole amide form nor the
benzylamide form under microwave or thermal heating
conditions. Therefore, our nickel-catalyzed reaction may not
proceed through Halberg’s proposed mechanism of direct
involvement of carbon monoxide from the decomposition
of DMF. Second, the influence of the electronic property of
the aryl substrate was clearly shown by the fact that the
reaction yielded an amide-coupled product. Aryl halides
bearing electron-donating groups showed higher reactivities
than aryl halides bearing electron-withdrawing groups, and
the latter produced dehalogenated arene as a major side
product. This implies that the pathway of â-hydride elimina-
tion is competing with the reductive elimination pathway to
produce amide coupling. Based on this reaction tendency
and side product formation, two competing pathways were
proposed, as shown in Scheme 1.
a
Reaction conditions: 1.0 equiv of aryl halides, 5 mol % of 1, 5 mol %
of Ni(OAc)2‚4H2O, 4 equiv of NaOMe, DMF (0.5 mL), 1,4-dioxane (3.0
mL). b Determined by GC. Isolated yields of compounds. Yield of
c
d
e
dehalogenated arene. 10 mol % of 1 and 10 mol % of Ni(OAc)2‚4H2O
were used.
whether the alkoxide-DMF adduct is formed first and then
coordinated to a nickel complex or whether nickel-
coordinated DMF is attacked by alkoxide. The electronic
properties of the aryl substrate are a primary means of
determining the pathway of this reaction. One pathway (path
A) is the nonproductive â-hydrogen elimination pathway,
using an electron-deficient nickel center with an electron-
withdrawing aryl group to favor nickel hydride formation
Both proposed paths proceed through the nickel-coordi-
nated alkoxide-DMF adduct 4, although it is not clear
Org. Lett., Vol. 9, No. 22, 2007
4617

Scheme 1. Proposed Mechanism
and subsequent production of the reduced arene. The second
yield obtained using DMF, but nevertheless implies that
1-formylpiperidine may be useful in catalytic aminocarbo-
nylation reactions. Further mechanistic elucidation and
studies of this aminocarbonylation protocol are in progress
in our laboratory.
pathway relies on a nickel amido intermediate (path B) to
give an amide product through reductive elimination. Al-
ternatively, this product might be produced by aryl group
insertion to nickel-coordinated DMF, followed by â-hydro-
gen elimination (path C). However, considering the strong
tendency of base species to yield amide products, the role
of the base is likely a nucleophilic attack on the carbonyl
carbon in DMF (path A) rather than simple absorption of
the HX on nickel complex (path C).
On the basis of mechanistic studies, we tested the
feasibility of the universal use of nickel-catalyzed aminocar-
bonylations. When employing 10 equiv of 1-formylpiperidine
as the DMF replacement, the coupled product was obtained
with 20% yield (Scheme 2). This is much lower than the
In summary, the first synthetically useful, convenient, and
cost-effective nickel-catalyzed aminocarbonylation is pre-
sented here. The relatively inexpensive catalyst and ligand
systems used in this reaction confirm this method’s superior-
ity over other existing processes.
Acknowledgment. This work was supported by the Korea
Research Foundation Grant funded by the Korean Govern-
ment (MOEHRD, Basic Research Promotion Fund) (KRF-
2
006-003-C00178).
Scheme 2. Aminocarbonylation of 1-Formylpiperidine
Supporting Information Available: Reaction procedures
and spectral and analytical data for reaction products. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL702058E
4618
Org. Lett., Vol. 9, No. 22, 2007