Palladium-N-heterocyclic carbene an efficient catalytic system for the carbonylative cross-coupling of pyridine halides with boronic acids

Article abstract of DOI:10.1016/j.tet.2006.11.008

Carbonylative cross-coupling of different pyridyl halides with various boronic acids was studied using catalytic systems constituted of N-heterocyclic carbene type ligands and palladium. These systems easily obtained in situ from the corresponding imidazolium salt and palladium acetate appear more efficient toward bromopyridines than catalysts based on hindered and basic alkylphosphines such as tricyclohexylphosphine. Their higher efficiency was also evidenced by coupling using chloro- or dichloropyridines and chloroquinolines, which practically do not react with catalytic systems based on phosphines.

Full text of DOI:10.1016/j.tet.2006.11.008

Tetrahedron 63 (2007) 682–689
Palladium-N-heterocyclic carbene an efficient catalytic
system for the carbonylative cross-coupling of pyridine
halides with boronic acids
*
ꢀ
Eddy Maerten, Mathieu Sauthier, Andre Mortreux and Yves Castanet
´
Ecole Nationale Superieure de Chimie de Lille, Unite de Catalyse et Chimie du Solide, UMR CNRS 8181,
´
USTL, BP 90108, 59652 Villeneuve d’Ascq, France
Received 26 September 2006; revised 31 October 2006; accepted 1 November 2006
Available online 21 November 2006
Abstract—Carbonylative cross-coupling of different pyridyl halides with various boronic acids was studied using catalytic systems consti-
tuted of N-heterocyclic carbene type ligands and palladium. These systems easily obtained in situ from the corresponding imidazolium salt
and palladium acetate appear more efficient toward bromopyridines than catalysts based on hindered and basic alkylphosphines such as
tricyclohexylphosphine. Their higher efficiency was also evidenced by coupling using chloro- or dichloropyridines and chloroquinolines,
which practically do not react with catalytic systems based on phosphines.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Suzuki reaction) provides a straightforward access to this
class of compounds.4
Different methodologies have been described to synthesize
a-aryl pyridyl ketone derivatives that are useful intermedi-
ates in the syntheses of various natural products and drugs.1
However, many of them have the disadvantage of using
expensive or unfriendly reagents,2 giving poor or moderate
yields3 or being only effective with specific substrates. In
this context, we have recently reported that the palladium
catalyzed three component cross-coupling of pyridine
halides, carbon monoxide, and boronic acid (carbonylative
The main drawback of this approach is the direct coupling
reaction (i.e., without carbon monoxide insertion) giving
biaryl products in particular with electron deficient aryl
halides such as pyridyl halides.5 Actually, from a mechanis-
tic point of view, the first step of the catalytic cycle is the
oxidative addition of the substrate on a Pd0 species leading
to the formation of a palladium aryl complex (Scheme 1).
This one can insert a CO molecule (path a) or react directly
R'COR
R'-R
[Pd]
RX
[Pd]
R'
COR
[Pd]
R'
R
Path a
Path b
[Pd] R
X
XB(OH)2
[P d]
X
COR
R'B(OH)2
CO
XB(OH)2
R'B(OH)2
Scheme 1.
* Corresponding author. Tel.: +33 320 434 927; fax: +33 320 436 585; e-mail: yves.castanet@ensc-lille.fr
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.11.008

E. Maerten et al. / Tetrahedron 63 (2007) 682–689
683
with boronic acid by transmetalation (path b). In the first
case the expected ketone is obtained via transmetalation
and reductive elimination whereas in the second case a biaryl
compound is formed. As the two paths (a and b) are in com-
petition, the electronic properties of the substrate, ligand of
the palladium as well as the reaction conditions can have
a great influence on the selectivity of the reaction.
and CO pressures were tested in combination with various
palladium precursors and 2 equiv of 1,3-bis(2,4,6-trimethyl-
phenyl)-imidazolium chloride (AHCl) in order to find the
best reaction conditions (Scheme 3).
+
+
N
In this context, we have shown that a proper choice of the re-
action conditions (CO pressure, palladium precursor, ligand,
solvent,.) allowed us to obtain high selectivity for the
desired pyridyl ketone.4 Another limitation of this reaction
arises from the fact that whilst iodides and bromides are
efficiently coupled under carbonylative conditions, chlo-
rides suffer from a lack of reactivity.
N
N
N
Cl
AHCl
Cl
BHCl
+
N
N
Cl
CHCl
Direct coupling of aryl chlorides, which are much cheaper
and more readily available compounds than their iodide or
bromide counterparts remained a challenge for a long time.
However, recently, remarkable progress has been achieved
in this field6 and different ligands (generally bulky electron
rich phosphanes or bulky carbenes associated to palladium
on the form of co-ordinatively unsaturated complexes) have
been reported as efficient catalysts for the direct coupling
(Suzuki, Stille, Heck.reactions) of even deactivated chlo-
rides under relatively mild conditions. Nevertheless there
are very few examples of activation of aryl chlorides under
CO conditions.7
Scheme 3. Imidazolium salt (Imd) precursors of carbenes A, B, and C.
Selected results are shown in Table 1. In polar aprotic sol-
vents such as DMF or NMP and with Pd(OAc)2 as palladium
source, no reaction took place (entries 4 and 5). Ethereal sol-
vents were more convenient and among them 1,4-dioxane
was the best solvent, which enabled complete conversion
of 1a, within reasonable reaction time and with high selec-
tivity, into 2-benzoylpyridine 4 (entry 2). With other ethers
such as anisole and THF, both the reactivity and selectivity
were slightly lower (entries 1 and 3). As in direct Suzuki
coupling reactions (without CO), the base used also greatly
affected the efficiency of the process,9 the best results were
obtained with alkaline carbonates and more specifically with
Cs2CO3 although with K2CO3 the conversion was only
slightly lower (compare entries 2 and 6). The use of KF,
which is also an effective base for direct coupling,10 led to
low conversion and the main product obtained was biphenyl,
which resulted from direct homocoupling. Finally, although
alkoxide anions are known to reduce imidazolium salts
into their corresponding carbenes,11 Mg(OEt)2 or t-BuOK
proved to be totally ineffective (entries 9 and 10).
We have previously reported that the use of a palladium-
catalyst based on N-heterocyclic carbene ligands allowed
the effective carbonylative coupling of different chloro-
pyridines with phenylboronic acid.8 As a part of an ongoing
project on the carbonylation of pyridyl halides, we have
studied the behavior of this class of carbene as a ligand in the
carbonylative cross-coupling of different haloazines with
different boronic acids and compared their efficiency in these
reactions with the use of tricyclohexylphosphine previously
found to be the most effective ligand for this reaction.4
2. Results and discussion
The CO pressure had little effect on the reaction since only
a slight decrease in the selectivity occurred on lowering the
pressure from 50 to 5 bar whereas the activity remained
unchanged (compare entries 2, 11, and 12).
2.1. Carbonylative cross-coupling of bromopyridines
As the use of carbene ligand in carbonylation reaction was
not well documented at the beginning of these works, we
first explored the reaction with 2-bromopyridine, 1a, and
phenylboronic acid (see Scheme 2). Various solvents, bases,
On the other hand, activity and selectivity were greatly
influenced by the nature of the catalytic precursor; using
[Pd(allyl)Cl]2 as precursor resulted in a remarkable enhance-
ment of the activity but to the slight detriment of the selec-
tivity. In contrast PdCl2 was less efficient than Pd(OAc)2.
Pd(OAc)2 - Imd
CO, ArB(OH)2
Base
+
X
COAr
Ar
N
N
N
1a : X = 2-Br; 1b : X = 2-Cl
2a : X = 3-Br; 2b : X = 3-Cl
3a : X = 4-Br; 3b : X = 4-Cl
4 : 2-COAr
5 : 3-COAr
6 : 4-COAr
Taking into account the catalytic performances of the Pd-
carbene ligand system, slightly better than those obtained
with PCy3 (compare Table 1, entries 3 and 15), we have
tested its behavior toward various bromoazines and different
boronic acids under the conditions previously optimized for
1a with the aim to delimit the scope and limitation of the
process and also for synthetic purpose.
X
COAr
N
Pd(OAc)2 - Imd
CO, ArB(OH)2
Base
Cl
N
Cl
7a : X = Br; 7b : X = Cl
8
Comparative results between the use of ligand A and PCy3
are summarized in Table 2. Starting with 2-bromopyridine
Scheme 2.

684
E. Maerten et al. / Tetrahedron 63 (2007) 682–689
Table 1. Palladium–imidazolium-catalyzed carbonylative coupling of 2-bromopyridine (1a) with phenylboronic acid
Entry
Catalyst precursor
Solvent
Base
P (CO) (bar)
Time (h)a
Conversion of 1a (%)
Selectivity of 4b (%)
1
2
3
4
5
6
7
8
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
PdCl2
Anisole
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
K2CO3
50
50
50
50
50
50
50
50
50
50
5
24
21
22
22
21
22
22
22
22
22
21
21
30
5.5
20
83
100
88
0
93
95
89
—
—
95
0
1,4-Dioxane
THF
DMF
NMP
0
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
1,4-Dioxane
THF
95
35
67
7
KF
K2CO3/CsCl
Mg(OEt)2
t-BuOK
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
K2CO3
94
0
9
10
11
12
13
14
15
0
—
91
94
96
90
90
100
100
68
99
74
10
50
50
5
[Pd(allyl)Cl]2c
PdCl2(PCy3)2
General conditions: 1a¼2.1 mmol, PhB(OH)2¼2.2 mmol, base¼4.0 mmol, [Pd]¼0.06 mmol, AHCl¼0.12 mmol, solvent¼15 mL, and T¼100 ꢀC.
a
Reaction time was not necessarily optimized.
Selectivity for carbonylated product 4; 2-phenylpyridine accounts for the balance.
Without AHCl.
b
c
and p-tolylboronic acid, high conversions were achieved
with both ligands. However, the results remained similar to
those obtained with phenylboronic acid with a slightly better
activity and selectivity for ligand A. In contrast, a remarkable
enhancement of the activity was observed with o-tolylbor-
onic acid using a catalyst based on ligand A. Actually, the
reaction went almost to completion only after 6 h while
more than 20 h were needed with PCy3 (or with A in the
case of phenylboronic acid). In the same way, the sterically
hindered 1-naphthylboronic acid led to a better activity than
PhB(OH)2. Other boronic acids such as p-methoxyphenyl,
p- and m-chlorophenylboronic acids exhibited a reactivity
similar or lower to the one observed with PhB(OH)2.
However, it is difficult to connect the activity with the elec-
tronic properties of the substituent of the aryl moiety of the
boronic acid. It is also noteworthy that in the test carried out
with m-chlorophenylboronic acid and carbene A (entry 5), a
large amount of chlorobenzene was found in the reaction
medium resulting from the decomposition of the boronic acid
(boronic acids bearing withdrawing groups are prone to
deboronation). Thus the modest conversion obtained in this
case was probably due to the lack of reagent. Finally, sodium
tetraphenylborate reacted in the same way as PhB(OH)2 and
provided a high yield of the corresponding ketone even using
0.25 equiv of NaB(Ph)4 perbromopyridine (entry 10).12
Under the same conditions as those used with 2-bromopyri-
dine, 3-bromopyridine was found to react much more slowly
with both ligands whichever boronic acid was used. This
observation is not surprising since the carbon–halogen
bond in the 3-position of the pyridine ring is known to be
deactivated toward the nucleophilic substitution1a and hence
toward the oxidative addition onto the palladium in contrast
with the 2-position. However, carbene A led to much better
results than PCy3 both in terms of conversion, particularly in
the case of m-chlorophenylboronic acid, and selectivity with
o- or p-tolylboronic acids.
Table 2. Carbonylative coupling of various bromoazines with different aryl
boronic acids
Entry Azine Boronic Ligand Solvent Time Conversion Selectivityc
acid Ar
(h) (%)
(%)
1a
2b
3a
4b
5a
6b
7a
8a
9a
1a
1a
1a
1a
1a
1a
1a
1a
1a
o-Tolyl
o-Tolyl
p-Tolyl
p-Tolyl
m-ClPh
m-ClPh PCy3 THF
p-ClPh
p-MeOPh A
A
PCy3 THF
Dioxane
6
22
97
83
99
100
74
100
100
87
100
100
70
100
86
45
83
48
70
96
95
97
82.5
93
50
95
96
75
82
88
67
86
49
85
51
92
71
72
47
90
76
A
Dioxane 24
PCy3 THF 40
Dioxane 30
62
A
A
Dioxane 21
Dioxane 45
Dioxane 15
Dioxane 22
Dioxane 46
Finally 4-bromo-7-chloroquinoline reacted very easily even
with the hindered o-tolylboronic acid since in this case the
reaction went almost to completion within only 4 h with the
two ligands (entries 21 and 22) but once again with a better
selectivity when using the carbene ligand. This example evi-
dences the enhanced reactivity of the halogen in 4-position
of the pyridinic ring toward this reaction and is in accor-
dance with results previously reported with iodopyridines.4
Naphthyl
NaB(Ph)4
Ph
A
A
A
10a 1a
11a 2a
12b 2a
13a 2a
14b 2a
15a 2a
16b 2a
17a 2a
18b 2a
19a 7a
20b 7a
21a 7a
22b 7a
PCy3 THF
Dioxane 30
PCy3 THF 30
Dioxane 30
PCy3 THF 30
Dioxane 30
30
62
o-Tolyl
o-Tolyl
p-Tolyl
p-Tolyl
m-ClPh
m-ClPh PCy3 THF
Ph
Ph
A
A
A
12
A Dioxane 4.5 75
PCy3 Dioxane 4.5 86
A Dioxane
PCy3 Dioxane
2.2. Carbonylative cross-coupling of chloropyridines
o-Tolyl
o-Tolyl
4
4
96
96
Having demonstrated the effectiveness of the ligand A in the
palladium catalyzed carbonylative cross-coupling of bromo-
pyridines with boronic acids, the applicability of this system
toward various chloropyridines was next examined (Table 3)
under the conditions optimized for bromopyridines (i.e.,
using 1,4-dioxane as solvent and Cs2CO3 as base).
General conditions: substrate¼2.1 mmol, boronic acid¼2.2 mmol except
entry 9: NaB(Ph)4¼0.55 mmol, PCO¼50 bar, T¼120 ꢀC, solvent¼15 mL.
a
Pd(OAc)2¼0.06 mmol, AHCl¼0.12 mmol, and Cs2CO3¼4 mmol.
b
PdCl2(PCy3)2¼0.06 mmol and K2CO3¼4 mmol.
c
Selectivity for the carbonylated product.

E. Maerten et al. / Tetrahedron 63 (2007) 682–689
Table 3. Carbonylative cross-coupling of 2-chloropyridine
685
80
60
40
20
0
0,04
0,03
0,02
0,01
0
Entry Liganda
Temp Time Conversion Selectivity
(ꢀC)
(h)
(%)
of 4b (%)
1c
2
3
4
5
6
7
8
A (2)
A (2)
A (2)
A (2)
A (2)
A (1)
A (1.5)
A (4)
140
100
120
140
150
140
140
140
1
40
40
22
24
30
22
30
30
22
60
22
30
30
22
100
35
63
72
60
32
67
49
30
88
40
72
48
39
73
g
—
86
93
90
87
84
89
87
92
91
92
92
89
90
91
conv (%)
9
A (2) preformed 140
10d
11e
12
13
14
15f
A (2)
A (2)
A (1)+PCy3(1)
B (2)
C (2)
140
140
140
140
140
140
mmol de trimethylaniline
0
10
20
Time (h)
30
40
A (2)
Conditions: 1b¼2.1 mmol, PhB(OH)2¼2.2 mmol, Cs2CO3¼4.0 mmol,
[Pd]¼0.06 mmol, dioxane¼15 mL, and PCO¼50 bar.
Figure 1. Carbonylative coupling of 2-chloropyridine; conversion and
decomposition of the catalytic system versus time. Conditions: See Table 3,
entry 4.
a
The number inside brackets refers to the ligand to Pd ratio.
Selectivity for carbonylated product 4.
b
c
Without CO.
Percolated dioxane.
d
e
system exhibited high activity (about 50% conversion after
4 h), but after several hours this dramatically dropped. In
parallel, GC–MS analyses of the reaction medium showed
the appearance of trimethylaniline during the course of the
reaction. The formation of this compound can only be
explained by the total decomposition of the carbene A (or
its imidazolium salt precursor). However, determination of
the quantity of trimethylaniline formed proved that only
a minor part of the carbene was retrieved in this form (about
10%) and that the rate of formation of the aniline decreased
with the reaction rate.
Addition of 0.2 mL of water.
PhB(OH)2¼4 mmol.
f
g
100% selectivity for 2-phenylpyridine.
At the same temperature (100 ꢀC) to that one used with 1a,
the carbonylative cross-coupling of 2-chloropyridine with
phenylboronic acid was very slow (only 30% conversion af-
ter 40 h, see Table 3, entry 2). Increasing the temperature led,
as expected, to a great increase in the activity with practi-
cally no effect on the selectivity. However, 140 ꢀC seems
to be optimal, leading to 72% conversion after 16 h. At
150 ꢀC both the overall conversion and the selectivity de-
creased slightly, at this temperature the catalytic system
probably reached its stability limit.
The decomposition of the boronic acid under the drastic
conditions used was also envisioned to explain the decrease
of the reaction rate, therefore, a run was carried out with
a higher quantity (2 equiv PhB(OH)2/chloropyridine) but
unfortunately no increase in the conversion was observed
(entry 15).
The influence of the palladium on ligand molar ratio was
also studied (with various amounts of imidazolium salt).
For a 1:1 ratio low conversion was observed whereas it is
noteworthy that this ratio was reported as leading to the
best results in the case of the direct Suzuki coupling of
aryl chlorides without CO.13 Ratios ranging between 1:1.5
and 1:2 led to the best activities. With a higher amount of
ligand the conversion markedly decreased. The latter obser-
vation is probably related to the fact that when higher
quantities of ligand are used, the coordination sphere of
the palladium center is too hindered to allow easy oxidative
addition and therefore good activity. On the contrary with
low amounts of ligand the lack of efficiency of the catalyst
is probably due to the fact that, on one hand, CO can act
as a ligand with poor sigma donor character and good p
acceptor properties, thus the electronic density of the metal
decreases and consequently the oxidative addition is more
difficult. This influence of the CO on the reaction rate is per-
fectly highlighted by an experiment carried out in the ab-
sence of CO but under the same conditions to those giving
the best conversion with CO. In this case, total conversion
into the coupling Suzuki type product was obtained in
only 1 h even with a A/Pd ratio of 2 (entry 1). On the other
hand, it is also possible that the system with low amount of
ligand A is not stable enough over a long period. Actually,
Figure 1 shows that at the beginning of the reaction the
In order to enhance the stability of the catalyst the use of
a mixture of carbene A and tricyclohexylphosphine,14 was
also checked but provided no improvement in the conversion
in comparison with A alone. In the same way, pre-formation
of the free carbene A in a separate vessel by action of a base
on the imidazolium salt and addition to the reaction medium
just before introduction of CO has a detrimental effect on the
conversion. Finally, bulky carbene B (with isopropyl substit-
uents on the phenyl groups) or C (the saturated counterpart
of A) formed in situ from the corresponding imidazolium
salts, was also tested but were found to be much less efficient
than A.
Taking into account these different results, we have been en-
deavoring to improve the long term stability of the catalyst.
In this context, we have studied the influence of the purity of
the solvent. Much better results in terms of conversion (88%
after 22 h) were obtained when dioxane was percolated just
before use through freshly activated alumina (entry 10). In
contrast addition of a small quantity of water (0.2 mL in
15 mL of solvent) had a very detrimental effect since the
conversion dropped to 40% after 60 h (entry 11). Better
defined palladium species than the one obtained in situ

686
E. Maerten et al. / Tetrahedron 63 (2007) 682–689
100
80
60
40
20
0
from Pd(OAc)2 and the imidazolium salts were also tested.
For this purpose, complexes I and II in which the metal is
coordinated to one or two carbenes B, were synthesized
according to literature methods15,16 (see Schemes 4 and 5).
R
R
Cl
Cl
+ 1/2
Pd
Pd
Pd
Complex I
Pd(OAc)2 + carbene A, 140°C
Pd(OAc)2 + carbene A, 120°C
THF, 1 h,rt
Cl
R
R
R = 1-(2,6-diisopropylphenyl)
Scheme 4.
I
0
5
10
15
20
Time (h)
25
30
35
With complex II at 140 ꢀC no reaction occurred until 20 h
after which a very low activity was observed (12% con-
version after 45 h) with concomitant formation of 2,6-di-
isopropylaniline. In contrast, complex I was a much more
active catalytic precursor even at a lower temperature
(120 ꢀC) since the conversion reached 85% after 22 h and
interestingly after that period the catalyst remained still
active to give an almost total conversion after 30 h. On the
other hand, Figure 2 shows that when the reaction was
carried out with complex I, low activity was observed in
the early stage of the reaction (lower activity than the one
obtained with the catalyst formed in situ from Pd(OAc)2+
AHCl at the same temperature). However, after this induc-
tion period, which probably corresponds to the time needed
to reduce the complex I into the catalytic species, the activity
greatly increased and started to decrease after about 5–6 h.
This decrease was however, much lower than the case of
catalyst formed in situ and thus the reaction can go to
completion.
Figure 2. Carbonylative coupling of 2-chloropyridine catalyzed by complex
I and systems formed in situ from Pd(OAc)2 and AHCl. Conditions: [Pd]¼
0.06 mmol, 1b¼2.1 mmol, PhB(OH)2¼2.2 mmol, Cs2CO3¼4.0 mmol,
[Pd]¼0.06 mmol, dioxane¼15 mL, PCO¼50 bar. Addition of AHCl in the
runs carried out with Pd(OAc)2.
conversion was obtained only after 4 h. This higher activity
of the Cl in the 2-position is probably due to the withdrawing
effect of its counterpart in the 3-position. On the other hand,
the influence of the 2-Cl (or 2-COPh) on the reactivity of the
3-position is limited since no coupling product at this posi-
tion was observed. Nevertheless a substantial amount of
reduction product i.e. 2-benzoylpyridine was also obtained.
Cl
Cl
Y
X
Pd(OAc)2 - AHCl
CO, PhB(OH)2
4h, 90% conv.
N
N
X = COPh, Y = Cl : 72%
X = Ph, Y = Cl : 13%
X = COPh, Y = H : 15%
The high efficiency of complex I in comparison with the one
of complex II indicates that the catalytic species contains
only one carbene ligand. The fact that diisopropylaniline
was found in the reaction medium when the reaction starts
to take place with complex II, is in accordance with this
hypothesis. Complex II becomes active only after losing
one carbene (the low activity observed under these con-
ditions probably results from the low concentration of the
active species due to a very high stability of complex II ver-
sus its dissociation into a monocarbene palladium species).
Scheme 6.
Consistent with the above results, in the case of the 2,5-di-
chloropyridine, only the 2-position of this substrate reacted
to give the corresponding 2-aroyl derivative (Scheme 7).
However, in this case the influence on the reactivity of the
halogen in the 5-position was very low and the conversion
is very close to that observed with 2-chloropyridine more-
over as in the above case the reduction product was also
observed.
2.3. Carbonylative cross-coupling of dichloropyridines
Cl
Y
Pd(OAc)2 - AHCl
The reactivity of various dihalopyridines was also tested
with a catalyst formed in situ from Pd(OAc)2 and the imida-
zolium salt precursor of carbene A.
CO, PhB(OH)2
23h, 83% conv.
N
Cl
N
X
X = COPh, Y = Cl : 86%
X = Ph,Y = Cl : 10%
X = COPh, Y = H : 4%
As it can be seen (Scheme 6) 2,3-dichloropyridine was found
to be much more reactive than 2-chloropyridine since 90%
Scheme 7.
R
R
R
R
R
R
R
R
Pd[(Po-tol)3]2
R
Pd(Po-tol)3
Pd
Toluene, rt
Toluene, rt
R
R = 1-(2,6-diisopropylphenyl)
II
Scheme 5.

E. Maerten et al. / Tetrahedron 63 (2007) 682–689
687
Although the two halogen atoms of 2,6-dichloropyridine are
activated this substrate exhibited a low reactivity (only 64%
conversion after 31 h) and the reaction products resulted
mainly from mono coupling with a low ketone versus direct
coupling product ratio (Scheme 8).
pyridyl aryl ketones with high selectivities. Moreover, the
use of these catalytic systems allowed us to extend the scope
of the reaction to the economically attractive but poorly
reactive chloro- or dichlorazines even if, with this class of
compounds, the reaction takes place only with substrates
in which the C–Cl bond is activated by the heteroatom. It
also appears that the nature of the imidazolium salt and its
ratio versus Pd is crucial with chloroderivatives to obtain
good results. In this way, the best results were observed
with well defined catalyst precursor Pd(allyl)BCl possessing
one carbene (1,3-di(2,6-diisopropylphenyl)imidazolin-2-
ylidene) perpalladium. This catalyst in comparison with
catalytic systems obtained in situ from imidazolium salts is
active at a lower temperature, remains active during all the
reaction time and leads to a higher selectivity. Nevertheless,
carbonylative cross-coupling of chloroderivatives appears to
be much more difficult than simple coupling without CO and
is still a challenge in the case of non-activated or deactivated
chloroaryls.
Pd(OAc)2 - AHCl
CO, PhB(OH)2
31h, 64% conv
Cl
N
Cl
Y
N
X
.
X = COPh, Y = Cl : 54%
X = Ph,Y = Cl : 32%
X = COPh, Y = Ph : 4%
X = COPh, Y = COPh : 9%
Scheme 8.
In contrast with its chloro-homologue, Scheme 9 shows that
2,6-dibromopyridine was more reactive than 2-bromopyri-
dine since total conversion occurred in only 10 h instead of
22 h with 2-bromopyridine. On the other hand, the diketone
was obtained in only moderate yield (69%) plus a notable
quantity of the monocarbonylative-direct coupling product.
This selectivity for the dicarbonylative coupling product
seems modest but results from two consecutive reactions
thus one can think that the selectivity of each coupling to
give ketone is about 83%, which is slightly lower than that
observed with 2-bromopyridine but is in accordance with
the fact that a withdrawing effect (Br or COPh) promotes
the oxidative addition but is detrimental to the CO insertion
(Scheme 9).4
4. Experimental
4.1. General remarks
NMR spectra were recorded on a AC-300 Bruker spectro-
meter at 23 ꢀC in CDCl3; chemical shifts are reported in
parts per million downfield from TMS and were determined
with reference to the residual 1H (d¼7.25) and 13C (d¼77.0)
solvent peaks. All coupling constants are reported in Hertz.
GLC analyses were performed on a Chrompack CP 9001
apparatus equipped with a flame ionization detector and
a CPSil 5CB (25 mꢁ0.32 mm ID, Chrompack) column.
MS and HRMS were performed on a JMS-700m Station
mass spectrometer (JEOL) with either electron impact
(70 eV) or chemical (CH4) ionization mode. Melting points
are uncorrected. All the carbonylative cross-coupling prod-
ucts described hereafter gave similar IR spectra (KBr
pellets) with n(C]O) in the range 1690–1640 cmꢂ1 and
Pd(OAc)2 - AHCl
CO, PhB(OH)2
10h, 100% conv.
Br
N
Br
Y
N
X
X = COPh, Y = Ph : 31%
X = COPh, Y = COPh : 69%
Scheme 9.
Finally with 3,4-dichloropyridine, once again the activated
Cl (i.e., the Cl in the 4-position) reacted more rapidly than
in the case of its homologue 4-chloropyridine to give the cor-
responding coupling products but once again with a modest
selectivity for the ketone over the direct coupling product.
As in the case of the other dichloropyridines some reduction
occurred even before coupling since a large amount of
3-chloropyridine was obtained (Scheme 10).
n(CN) in the range 1590–1550 cmꢂ1
.
The commercially available Pd(OAc)2 (Acros), Cs2CO3
(Acros), PCy3 (Acros), N-heterocyclic imidazolium salts
(AHCl and BHCl from Strem, CHCl from Acros),
PhB(OH)2 (Acros), NaBPh4 (Acros), and dihalopyridines
(except 3,4-dichloropyridine) were used as received. Mono-
halopyridines were generally distilled in presence of KOH
and kept under a nitrogen atmosphere in a Schlenk tube.
1,4-Dioxane was percolated on activated alumina before
being degassed and conserved in Schlenk tube. The
substrates: 3,4-dichloropyridine,17 4-bromo-7-chloroquino-
line,18 and substituted aryl boronic acids19 and the com-
Cl
X
Cl
Y
Pd(OAc)2 - AHCl
CO, PhB(OH)2
7h, 65% conv.
N
N
X = COPh, Y = Cl : 47%
X = Ph, Y = Cl : 16%
X = COPh, Y = H : 13%
X = H, Y = Cl: 24%
plexes [Pd(allyl)Cl]2,20 [Pd(allyl)BCl],15 and PdB2 were
16
Scheme 10.
prepared according to literature procedures.
4.2. Typical procedure for carbonylative coupling
3. Conclusion
In a typical experiment (Table 3, entry 4), a 60-mL stainless
steel autoclave equipped with a magnetic stirring bar was
charged with all of the solid reagents i.e. Cs2CO3 (1.3 g,
4.0 mmol), Pd(OAc)2 (13.5 mg, 0.06 mmol), the imidazo-
lium salt (0.12 mmol), and PhB(OH)2 (268 mg, 2.2 mmol).
After flushing the atmosphere with N2, a solution of 1b
In summary, we have shown that imidazolium salts,
precursors of N-heterocyclic ligands, in combination with
palladium acetate provide efficient catalytic systems for
carbonylative cross-coupling reactions of pyridyl bromides
with different aryl boronic acids leading to a variety of

688
E. Maerten et al. / Tetrahedron 63 (2007) 682–689
(200 mL, 2.1 mmol) in 1,4-dioxane (15 mL) was added
under N2. The autoclave was flushed with CO, pressurized
to 50 bar and heated to 140 ꢀC. The reaction was monitored
by GC using dodecane as an internal standard. At the end
of the reaction, after cooling to room temperature, CH2Cl2
(15 mL) was added to the reaction mixture, the solution
was then filtered and the precipitate was washed with
CH2Cl2. The different filtrates were combined and the sol-
vent removed under vacuum. The residue was then purified
by silica gel chromatography to give analytically pure (1H,
13C NMR, and MS) ketone products.
J¼4.2 Hz, H-6), 8.05 (d, 2H, H-9 and H-3), 7.90 (t, 1H,
J¼7.5 Hz, H-4), 7.40–7.50 (m, 2H, H-5 and H-10).
13C{1H} NMR (CDCl3) d¼192.3, 154.6, 148.5, 139.3,
139.1, 134.5, 132.4, 128.4, 126.4, 124.6.
4.2.7. 1-Naphthyl-2-pyridylmethanone. White solid, mp¼
43–44 ꢀC. 1H NMR (CDCl3) d¼8.70 (d, 1H, J¼4.4 Hz,
H-6), 8.25 (m, 1H, H-15), 8.17 (d, 1H, J¼7.9 Hz, H-3),
8.03 (d, 1H, J¼8.2 Hz), 7.84–7.94 (m, 2H), 7.72 (d,
J¼7.0 Hz), 7.45–7.60 (m, 4H). 13C{1H} NMR (CDCl3)
d¼196.5, 155.4, 149.1, 137.0, 134.7, 133.8, 132.2, 131.2,
129.9, 128.4, 127.4, 126.5, 126.3, 125.6, 124.6, 124.1.
4.2.1. 2-Benzoylpyridine. White solid, mp¼42 ꢀC. 1H NMR
(CDCl3) d¼8.70 (d, 1H, J¼4.4 Hz, H-6), 8.04 (d, 1H,
J¼7.6 Hz, H-3), 8.01 (d, 2H, J¼7.5 Hz, H-9), 7.86 (td, 1H,
J¼7.6 and 1.7 Hz, H-4), 7.56 (t, 1H, J¼7.6 Hz, H-11),
7.44–7.49 (m, 3H, H-5 and H-10). 13C{1H} NMR (CDCl3)
d¼193.9 (CO), 155.3, 148.6, 137.1, 136.2, 132.9, 131.0,
128.2, 126.2, 124.6. MS (EI): m/z¼183 (M+, 38), 182 (62),
155 (M+ꢂCO, 92), 154 (42), 127 (6), 106 (M+ꢂPh, 10),
105 (M+ꢂPy, 90), 78 (Py+, 28), 77 (Ph+, 100), 52 (18), 51
(C4H+3, 76), 50 (28).
4.2.8. 3-Benzoylpyridine. White solid, mp¼40–41 ꢀC.
1H NMR (CDCl3) d¼9.01 (d, 1H, J¼2.0 Hz, H-2), 8.85
(dd, 1H, J¼5.0 and 2.0 Hz, H-6), 8.15 (dt, 1H, J¼8.0 and
2.0 Hz, H-4), 7.84 (d, 2H, J¼7.5 Hz, H-9), 7.75 (dd, 1H,
J¼8.0 and 5.0 Hz, H-5), 7.35–7.75 (m, 3H, H-10 and
H-11). 13C{1H} NMR (CDCl3) d¼194.8, 152.8, 151.0,
137.1, 136.6, 133.1, 129.9, 128.5, 128.4, 123.3. MS (EI):
m/z¼183 (M+, 96), 182 (48), 155 (M+ꢂCO, 8), 154 (10),
127 (5), 106 (M+ꢂPh, 40), 105 (M+ꢂPy, 100), 78 (Py+,
56), 77 (Ph+, 84), 63 (2), 51 (C4H+3, 78).
4.2.2. o-Tolyl-2-pyridylmethanone. Pale yellow solid,
mp¼67–68 ꢀC. 1H NMR (CDCl3) d¼8.67 (d, 1H, J¼
4.2 Hz, H-6), 8.05 (d, 1H, J¼7.8 Hz, H-3), 7.84 (td, 1H,
J¼7.8 and 1.5 Hz, H-4), 7.21–7.45 (m, 5H, H-5, H-10,
H-11, H-12, and H-13), 2.37 (s, 3H, H-14). 13C{1H} NMR
(CDCl3) d¼197.3 (CO), 154.9, 149.1, 137.7, 137.3, 136.9,
131.0, 130.9, 129.9, 126.4, 125.0, 124.0, 20.4. MS (EI):
m/z¼197 (M+, 9), 182 (M+ꢂCH3, 13), 168 (100), 119
(M+ꢂPy, 8), 91 (Tol+, 22), 78 (Py+, 2), 51 (C4H+3, 3).
1
4.2.9. o-Tolyl-3-pyridylmethanone. Yellow oil, H NMR
(CDCl3) d¼8.89 (s, 1H, H-2), 8.72 (t, 1H, J¼3.4 Hz, H-6),
8.05 (d, 1H, J¼7.8 Hz, H-4), 7.34–7.39, (m, 2H), 7.18–
7.29 (m, 3H), 2.32 (s, 3H, H-14). 13C{1H} NMR (CDCl3)
d¼196.4, 152.9, 151.0, 136.9, 136.8, 136.7, 132.9, 131.0,
130.7, 128.6, 125.1, 123.1, 19.8. MS (EI): m/z¼197 (M+,
27), 168 (30), 119 (M+ꢂPy, 10), 91 (Tol+, 26), 78 (Py+, 6),
51 (C4H+3, 5).
4.2.3. p-Tolyl-2-pyridylmethanone. Yellow solid, mp¼43–
4.2.10. p-Tolyl-3-pyridylmethanone. White solid, mp¼76–
77 ꢀC. 1H NMR (CDCl3) d¼8.96 (s, 1H, H-2), 8.78 (d, 1H,
J¼3.0 Hz, H-6), 8.08 (d, 1H, J¼6.7 Hz), 7.69 (d, 2H,
J¼7.5 Hz), 7.42 (d, 1H, J¼4.8 Hz), 7.28 (d, 2H, J¼
7.3 Hz), 2.43 (s, 3H, H-14). 13C{1H} NMR (CDCl3)
d¼194.4, 152.4, 150.6, 144.1, 137.1, 134.0, 130.2, 129.2,
123.3, 21.6. MS (EI): m/z¼197 (M+, 67), 182 (M+ꢂCH3,
100), 119 (M+ꢂPy, 51), 91 (Tol+, 52), 78 (Py+, 8), 51
(C4H+3, 7).
1
44 ꢀC. H NMR (CDCl3) d¼8.68 (d, 1H, J¼3.0 Hz, H-6),
7.81–7.98 (m, 4H, H-3, H-9, H-4, and H-13), 7.40 (t, 1H,
J¼7.2 Hz, H-5), 7.24 (d, 2H, J¼11.9 Hz, H-10 and H-12),
2.40 (s, 3H, H-14). 13C{1H} NMR (CDCl3) d¼193.4,
152.2, 148.3, 143.6, 136.8, 133.4, 130.9, 128.7, 125.8,
124.3, 21.6. MS (EI): m/z¼197 (M+, 29), 182 (M+ꢂCH3,
46), 169 (M+ꢂCO, 100), 119 (44), 91 (Tol+, 65), 78
(Py+, 3), 51 (C4H+3, 6).
4.2.4. m-Chlorophenyl-2-pyridylmethanone. Yellow oil,
1H NMR (CDCl3) d¼8.70 (d, 1H, J¼4.2 Hz, H-6), 8.03–
8.06 (m, 2H, H-3 and H-9), 7.85–7.97 (m, 2H, H-4 and
H-13), 7.48–7.56 (m, 2H, H-5, H-11), 7.41 (t, 1H,
J¼7.8 Hz, H-12). 13C{1H} NMR (CDCl3) d¼192.3, 154.3,
148.6, 137.8, 137.2, 134.2, 132.7, 130.9, 129.4, 129.1,
126.5, 124.7. MS (EI): m/z¼217 (M+, 24), 189 (M+ꢂCO,
100), 182 (M+ꢂCl, 31), 154 (M+ꢂCOꢂCl, 75), 139
(M+ꢂPy, 31), 111 (PhCl+, 45), 78 (Py+, 5), 77 (Ph+, 3), 75
(32), 51 (C4H+3, 10).
4.2.11. m-Chlorophenyl-3-pyridylmethanone. Yellow oil,
1H NMR (CDCl3) d¼8.96 (s, 1H, H-2), 8.80 (d, 1H,
J¼3.7 Hz, H-6), 8.08 (d, 2H, J¼7.8 Hz), 7.75 (s, 1H, H-9),
7.51–7.64 (m, 2H), 7.39–7.47 (m, 2H). 13C{1H} NMR
(CDCl3) d¼193.2, 152.8, 150.5, 138.1, 137.3, 134.9,
133.0, 132.6, 129.9, 129.7, 128.0, 123.5. MS (EI): m/z¼
217 (M+, 2), 182 (M+ꢂCl, 100), 139 (M+ꢂPy, 21), 111
(PhCl+, 18), 106 (M+ꢂPhCl, 6), 78 (Py+, 10), 51 (C4H+3, 7).
4.2.12. 4-Benzoylpyridine. White solid, mp¼70–71 ꢀC.
1H NMR (CDCl3) d¼8.81 (d, 2H, J¼3.9 Hz, H-2), 7.78 (d,
2H, J¼7.3 Hz, H-7), 7.49–7.67 (m, 5H, J¼7.3 Hz, H-3,
H-8, and H-9). 13C{1H} NMR (CDCl3) d¼195.1, 150.3,
144.3, 135.8, 133.5, 130.1, 128.6, 122.9. MS (EI):
m/z¼183 (M+, 34), 106 (M+ꢂPh, 10), 105 (M+ꢂPy, 100),
78 (Py+, 12), 77 (Ph+, 52), 51 (C4H+3, 38).
4.2.5. p-Methoxyphenyl-2-pyridylmethanone. Pale yellow
oil, 1H NMR (CDCl3) d¼8.70 (d, 1H, J¼4.4 Hz, H-6), 8.12
(d, 1H, J¼8.8 Hz, H-9), 7.98 (d, 1H, H-3, J¼7.8 Hz), 7.88
(m, 1H, H-4), 7.46 (m, 1H, H-5), 6.98 (d, 1H, H-10), 3.88
(s, 3H, CH3). 13C{1H} NMR (CDCl3) d¼192.2, 163.5,
155.7, 148.3, 136.9, 133.4, 128.9, 125.8, 124.5, 113.4, 55.4.
4.2.13. Phenyl-4-(7-chloroquinolyl)methanone. Pale solid,
mp¼104–105 ꢀC. 1H NMR (CDCl3) d¼9.01 (d, 1H,
J¼4.4 Hz, H-2), 8.19 (d, 1H, J¼1.95 Hz, H-5), 7.84 (d,
4.2.6. p-Chlorophenyl-2-pyridylmethanone.21 Pale yellow
1
solid, mp¼60–61 ꢀC. H NMR (CDCl3) d¼8.70 (d, 1H,

E. Maerten et al. / Tetrahedron 63 (2007) 682–689
689
Liebeskind, L. S., Ed.; JAI: London, 1998; Vol. 6, pp 187–
243; (d) Suzuki, A. Metal-Catalyzed Cross-Coupling Reac-
tions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH: New
York, NY, 1998; Chapter 2.
2H, J¼7.3 Hz, H-13), 7.82 (s, 1H, H-8), 7.65 (m, 1H, H-15),
7.50 (m, 2H, H-14), 7.47 (m, 1H, H-6), 7.41 (d, 1H,
J¼4.4 Hz, H-3). 13C{1H} NMR (CDCl3) d¼195.5, 150.5,
149.1, 144.2, 136.4, 136.0, 134.3, 130.3, 128.9, 128.8,
128.7, 126.7, 123.3, 119.8. MS (EI): m/z¼267 (M+, 72),
232(M+ꢂCl, 100), 105 (PhCO+, 63), 77 (Ph+, 48), 51
(C4H+3, 13).
6. For a recent review see: Littke, A. F.; Fu, G. C. Angew. Chem.,
Int. Ed. 2002, 41, 4176–4211.
7. For examples in alkoxycarbonylation see: (a) Ben-David, Y.;
Portnoy, M.; Milstein, D. J. Am. Chem. Soc. 1989, 111,
€
8742–8744; (b) Magerlein, W.; Indolese, A. F.; Beller, M.
Angew. Chem., Int. Ed. 2001, 40, 2856–2859; (c) Magerlein,
W.; Indolese, A. F.; Beller, M. J. Organomet. Chem. 2002,
641, 30–40.
4.2.14. o-Tolyl-4-(7-chloroquinolyl)methanone. White
1
solid, mp¼78–79 ꢀC. H NMR (CDCl3) d¼8.95 (dd, 1H,
€
J¼3.7 and 2.0 Hz, H-2), 8.80 (d, 1H, J¼2.0 Hz, H-5),
8.00–8.10 (m, 3H), 7.70 (d, 1H, J¼8.8 Hz), 7.07–7.42 (m,
3H, H-15), 2.51 (s, 3H, H-18). 13C{1H} NMR (CDCl3)
d¼197.5, 151.0, 150.5, 135.6, 132.4, 131.9, 131.6, 128.8,
128.6, 128.5, 128.1, 127.3, 126.6, 125.5, 121.0, 120.6,
21.1. MS (EI): m/z¼281 (M+, 100), 264 (40), 252
(M+ꢂCO, 72), 246 (M+ꢂCl, 45), 217 (35), 151 (TolCO+,
29), 119 (M+ꢂTolCO, 15), 91 (Tol+, 36).
8. Preliminary communication: Maerten, E.; Hassouna, F.;
Couve-Bonnaire, S.; Carpentier, J.-F.; Mortreux, A.;
Castanet, Y. Synlett 2003, 1874–1876.
9. For example see: (a) Ishiyama, T.; Murata, M.; Miyaura, N.
J. Org. Chem. 1995, 60, 7508–7510; (b) Norrild, J. C.;
Eggert, H. J. Am. Chem. Soc. 1995, 117, 1479–1484.
10. (a) Littke, A. D.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000,
122, 4020–4028; (b) Yin, J.; Ranka, M. P.; Zhang, X.-X.;
Buchwald, S. L. J. Am. Chem. Soc. 2002, 124, 1162–1163.
11. Arduengo, A. J., III; Krafczyk, R.; Schmutzler, R.; Craig,
H. A.; Goerlich, J. R.; Marshall, W. J.; Unverzagt, M.
Tetrahedron 1999, 55, 14523–14534.
References and notes
1. (a) Yates, F. S. Comprehensive Heterocyclic Chemistry;
Boulton, A. J., Mckillop, A., Eds.; Pergamon: Oxford, 1984;
Vol. 2, Chapter 2.08; (b) Balasubramanian, M.; Keay, J. G.
Comprehensive Heterocyclic Chemistry II; Mckillop, A., Ed.;
Pergamon: Oxford, 1995; Vol. 5, Chapter 5.06.
€
ꢀ
€
ꢀ
12. Skoda-Foldes, R.; Szekvolgyi, Z.; Kollar, L.; Berente, Z.;
ꢀ
Horvath, J.; Tuba, Z. Tetrahedron 2000, 56, 3415–3418.
13. Zhang, C.; Huang, J.; Trudell, M. L.; Nolan, S. P. J. Org. Chem.
1999, 64, 3804–3805.
2. (a) Hlasta, D. J.; Court, J. J. Tetrahedron Lett. 1989, 30, 1773–
ꢁ
€
14. (a) Herrmann, W. A.; Bohm, V. P. W.; Gstottmayr, C. W. K.;
Grosche, M.; Reisinger, C. P.; Weskamp, T. J. Organomet.
€
1776; (b) Gros, P.; Fort, Y.; Caubere, P. J. Chem. Soc., Perkin
Trans. 1 1997, 3597–3600; (c) Kondo, Y.; Shilai, M.;
Uchiyama, M.; Sakamoto, T. J. Chem. Soc., Perkin Trans. 1
1996, 1781–1782; (d) Choudary, B. M.; Bhuma, V.; Narender,
N. Indian J. Chem., Sect. B 1996, 35B, 281–282; (e)
Mohammadpoor-Baltork, I.; Sadeghi, M. M.; Mahmoodi, N.;
Kharamesh, B. Indian J. Chem., Sect. B 1997, 36B, 438–441;
(f) Mohammad-Baltork, I.; Hajipour, A. R.; Mohammadi, H.
Bull. Chem. Soc. Jpn. 1998, 71, 1649–1653; (g) Hajipour,
A. R.; Mahboobkhak, N. Indian J. Chem., Sect. B 1998, 37B,
285–287; (h) Hajipour, A. R.; Mohammadpoor-Baltork, I.;
Kianfar, G. Bull. Chem. Soc. Jpn. 1998, 71, 2655–2659.
3. (a) Sakamoto, T.; Kondo, Y.; Murata, N.; Yamanaka, H.
Tetrahedron Lett. 1992, 33, 5373–5374; (b) Sakamoto, T.;
Kondo, Y.; Murata, N.; Yamanaka, H. Tetrahedron 1993, 49,
9713–9720; (c) Hermann, C. K. F.; Sachdeva, Y. P.; Wolfe,
J. F. J. Heterocycl. Chem. 1987, 24, 1061–1065.
4. (a) Couve-Bonnaire, S.; Carpentier, J.-F.; Mortreux, A.;
Castanet, Y. Tetrahedron Lett. 2001, 42, 3689–3691; (b)
Couve-Bonnaire, S.; Carpentier, J.-F.; Mortreux, A.;
Castanet, Y. Tetrahedron 2003, 59, 2793–2799.
5. (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457–2483;
(b) Suzuki, A. J. Organomet. Chem. 1999, 576, 147–168;
(c) Miyaura, N. Advances in Metal-Organic Chemistry;
ꢁ
Chem. 2001, 617–618, 616–628; (b) Calo, V.; Giannoccaro,
P.; Nacci, A.; Monopoli, A. J. Organomet. Chem. 2002, 645,
152–157.
15. Viciu, M. S.;Germaneau, R. F.;Navarro-Fernandez, O.;Stevens,
E. D.; Nolan, S. P. Organometallics 2002, 21, 5470–5472.
16. Bohm, V. P. W.; Gstottmayr, C. W. K.; Weskamp, T.;
Herrmann, W. A. J. Organomet. Chem. 2000, 595, 186–190.
17. Marzi, B.; Bigi, A.; Schlosser, M. Eur. J. Org. Chem. 2001,
1371–1376.
18. (a) Schlosser, M.; Cottet, F. Eur. J. Org. Chem. 2002, 4181–
4184; (b) Peck, R. M.; Preston, R. K.; Creech, H. J. J. Am.
Chem. Soc. 1959, 81, 3984–3989.
19. (a) Wong, K.-T.; Chien, Y.-Y.; Liao, Y.-L.; Lin, C.-C.; Chou,
M.-Y.; Leung, M.-K. J. Org. Chem. 2002, 67, 1041–1044; (b)
Li, W.; Nelson, D. P.; Jensen, M. S.; Hoerrner, R. S.; Cai, D.;
Larsen, R. D.; Reider, P. J. J. Org. Chem. 2002, 67, 5394–5397.
20. Tatsuno, Y.; Yoshida, T.; Otsuka, S. Inorganic Synthesis;
Angelici, R. J., Ed.; Reagents for Transition Metal Complex
and Organometallic Syntheses; Wiley: New York, NY, 1990;
Vol. 28, p 343.
21. Yin, Z.; Zhang, Z.; Kadow, J. F.; Meanwell, N. A.; Wang, T.
J. Org. Chem. 2004, 69, 1364–1367.