Mechanism of the copper-free palladium-catalyzed sonagashira reactions: Multiple role of amines

Article abstract of DOI:10.1002/chem.200600574

Amines used as bases in copper-free, palladium-catalyzed Sonogashira reactions play a multiple role. The oxidative addition of iodobenzene with [Pd⁽⁾(PPh₃)₄] is faster when performed in the presence of amines (piperidine>m

Full text of DOI:10.1002/chem.200600574

DOI: 10.1002/chem.200600574
Mechanism of the Copper-Free Palladium-Catalyzed Sonagashira Reactions:
Multiple Role of Amines
Asma Tougerti, Serge Negri, and Anny Jutand*[a]
Abstract: Amines used as bases in
copper-free, palladium-catalyzed Sono-
gashira reactions play a multiple role.
The oxidative addition of iodobenzene
posed for Sonogashira reactions, de-
pending on the ligand and the amine.
When L=PPh3, its substitution by the
its substitution in trans-[PdI(Ph)-
(AsPh3)2]by the piperidine is easier
than that by the alkyne, leading to a
different mechanism: substitution of
AsPh3 by the amine is followed by sub-
stitution of the second AsPh3 by the
AHCTREUNG
amine in trans-[PdI(Ph)(PPh3)2]is less
R
with [Pd0
A
favored than that of the alkyne. A
mechanism involving prior coordina-
tion of the alkyne is suggested, fol-
lowed by deprotonation of the ligated
alkyne by the amine. When L=AsPh3,
formed in the presence of amines (pi-
peridine>morpholine). Amines also
alkyne to generate [PdI(Ph)
A
substitute one ligand
L
in trans-
ACHTREUNG
[PdI(Ph)(L)2](L =PPh3,
AsPh3)
alkyne by an external amine leads to
the coupling product. This explains
why the catalytic reactions are less effi-
cient with AsPh3 than with PPh3 as
ligand.
formed in the oxidative addition. This
reversible
[PdI(Ph)L
reaction,
which
gives
Keywords: alkynes
·
amines
·
ACHTREUNG
palladium · reaction mechanisms ·
Sonogashira reaction
order AsPh3 >PPh3 and piperidine>
morpholine. Two mechanisms are pro-
Introduction
complex 3, which is able to undergo a reductive elimination,
affording the coupling product and the Pd0 catalyst.
Amines are often used as bases in copper-free, palladium-
catalyzed Sonogashira reactions [Eq. (1)].[1] Specific amines
are required, such as the secondary amines piperidine,[1a,d,f,g]
morpholine,[1h] and diisopropylamine.[1f,h] The amines are
usually used in large excess,[1c,d,f,g] or as the solvent.[1a,d,e]
In a preliminary study, we established that amines (for ex-
ample, piperidine, morpholine, diisopropylamine) react with
trans-[PdX(Ar)
ACHTREUNG
[PdX(Ar)
A
ACHTREUNG
The mechanism of the copper-free Sonogashira reaction is
not known. The mechanism proposed in Scheme 1 is derived
from a related one.[1h] The key step is a complexation of the
alkyne to the complex [PdX(Ar)(L)2] (1) formed in the oxi-
dative addition, with release of one ligand L. This generates
[PdX(Ar)
(PPh3)(h2-alkyne)]( 2), in which the proton of the
R
ligated alkyne is more acidic than in the free alkyne.[2] De-
protonation of the ligated alkyne by the amine generates
[a]A. Tougerti, S. Negri, Dr. A. Jutand
École Normale SupØrieure
DØpartement de Chimie, UMR CNRS-ENS-UPMC 8640
24 Rue Lhomond, 75231 Paris Cedex 5 (France)
Fax : (+33)144-323-325
E-mail: Anny.Jutand@ens.fr
Scheme 1.
666
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676

FULL PAPER
equilibrium constants K have been calculated [Eq. (2)].[3]
The reversible substitution of one PPh3 group by the amine
may thus compete with the substitution of PPh3 by the
formed at room temperature with the amine as solvent. For
comparison, the reagents and catalyst concentrations were
the same as those of the reaction catalyzed by [Pd0-
alkyne to generate [PdX(Ar)
A
A
step in the postulated mechanism of the copper-free Sonoga-
shira reaction (Scheme 1). The fact that the amine is often
used in large excess,[1c,d,f,g] or even as solvent,[1a,d,e] may favor
the substitution of the phosphine by the amine.
Table 1. Palladium-catalyzed Sonogashira reactions[a] (Scheme 4).
Entry
L
Amine
Recovered
PhI [%][b]
Yield
of 6 [%][b]
1
2
3
4
PPh3
AsPh3
PPh3
piperidine
piperidine
morpholine
morpholine
0
47
40
45
94
44
56
50
AsPh3
[a]PhI (1mmol) and 5 (2 mmol) in amine (piperidine or morpholine;
3 mL). [b]Yields are relative to the initial PhI. They were determined in
1
the crude after work-up, by H NMR spectroscopy using CHCl2CHCl2 as
an internal standard.
In other work related to the palladium-catalyzed Stille re-
ꢀ
action (cross-coupling between iodobenzene and [CH2=CH
SnBu3]), it was established that AsPh3 was a more efficient
ligand than PPh3.[4] This was rationalized by the fact that
the same reaction time of 3 h15 min, to compare the effi-
ciency of the ligand AsPh3 with that of PPh3, as well as the
efficiencies of the amines.
From the results of the catalytic reactions reported in
Table 1, PPh3 appears to be a better ligand than AsPh3 for
the palladium-catalyzed Sonogashira reaction, whatever the
base (piperidine or morpholine). Piperidine was much more
efficient than morpholine when PPh3 was considered (en-
tries 1, 3), whereas similar results were obtained for AsPh3
whatever the base (entries 2, 4).
ꢀ
AsPh3 was more easily substituted by CH2=CH SnBu3 in
trans-[PdI(Ph)
plexes [PdI(Ph)
trans-[PdI(Ph)
(PPh3)2] (1a).[4,5] In other words, AsPh3 is a
A
2
(AsPh3)(h -CH2=CH SnBu3)]than PPh in
ACHTREUNG
more labile ligand than PPh3 in trans-[PdI(Ph)(L)2]com-
plexes.
This encouraged us to investigate the role of ligands and
amines in copper-free Sonogashira reactions more deeply.
We report herein that AsPh3 is less efficient than PPh3 in
the catalytic reactions. The amine may be involved in three
different steps of the catalytic cycle and two different mech-
anisms may operate according to the nature of the ligand
and amine.
A
N
AHCTREUNG
AHCTREUNG
AHCTREUNG
dative addition with PhI.[7] This indicates that the oxidative
addition is not the rate-determining step of the Sonogashira
reactions investigated here. A main difference between the
two precursors is the concentration of free PPh3, which is
Results and Discussion
Comparative efficiency of PPh3 and AsPh3 in copper-free
palladium-catalyzed Sonogashira reactions: The effect of the
ligand PPh3 or AsPh3 was probed in the reaction of PhI with
the alkyne 5 [Eq. (3)].
more important when starting from [Pd0
[Pd0(PPh3)3]+PPh3) than from [Pd0
[Pd0 (PPh3)2]+dba; no free PPh3).[7] This implies that
(dba)
ACHTREUNG
A
ACHTREUNG
A
ACHTREUNG
PPh3 is involved in the rate-determining step of the catalytic
cycle by a decelerating effect. In other words, the rate-deter-
mining step involves the release of PPh3 by a ligand substi-
tution process. The fact that the catalytic reaction works
better with piperidine than with morpholine when L=PPh3
suggests that the amine also is involved in the rate-deter-
mining step. At the least, we need to understand why AsPh3
is much less efficient than PPh3, when piperidine is used as
base and solvent.
Such a reaction catalyzed by [Pd0
Linstrumelle et al., using piperidine as solvent (5% [Pd0-
(PPh3)4] , 258C, 6 h, 96% yield).[1a,6] We have used [Pd0-
(dba)2](dba =dibenzylidene acetone) as the catalytic pre-
(PPh3)4]was reported by
C
Reaction of amines with trans-[PdI(Ph)
formed in oxidative addition: The reaction of PhI with [Pd0-
(dba)2]associated with two equivalents PPh gives trans-
A
ACHTREUNG
cursor, in order to vary the nature of the ligand more easily.
First we checked that no side reaction occurred between the
amine (morpholine or piperidine) and the dba which would
ACHTREUNG
3
[PdI(Ph)
amines [Eq. (2)].[3] The reaction of PhI with [Pd0
sociated with two equivalents AsPh3 gives trans-[PdI(Ph)-
(AsPh3)2] (1b) (Scheme 2).[8]
ACHTREUNG
AHCTREUNG
be released from the precursor [Pd0
ACHTREUNG
AHCTREUNG
Chem. Eur. J. 2007, 13, 666 – 676
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
667

A. Jutand et al.
chloroform (0.75 mL), the solution turned from dark orange
to yellow. The signals of 7b disappeared partially to give the
new signals mentioned in Table 2. Morpholine (3 equiv) was
required to consume 7b completely. Therefore, morpholine
reacted more easily with the dimer 7b to generate the new
complex 4bm, still in a reversible reaction. This new com-
plex was crystallized and its formula deduced from its X-ray
structure (Figure 1 (top) and Tables 3 and 4), attesting to
the substitution of one AsPh3 by one morpholine ligand,
which is in a trans position relative to the remaining AsPh3
ligand.
Scheme 2.
As previously established, trans-[PdI(Ph)
in equilibrium with the dimeric complex [PhPd
(AsPh3)]2 (7b) in chloroform [Eq. (4)].[5]
ACHTREUNG
AHCTREUNG
ACHTREUNG
When morpholine (10 mmol) was added to an NMR tube
containing trans-[PdI(Ph)(AsPh3)2] (1b; 10 mmol) in CDCl3
AHCTREUNG
(0.75 mL), the NMR signals of the protons of the Ph group
of 1b disappeared partially to give new signals (Table 2) the
Table 2. 1H NMR (400 MHz, CDCl3, 258C, TMS) spectrum of 4bm (see
Figure 1, bottom, for the attribution of protons).
Proton
d
H1
H’4
H’1
H3
H4
H’3
H’2
H2
2.82 (d, d, d, d, J=12.8, 12.8, 12.8, 3.7 Hz, 1H)
2.99 (d, J=12.8 Hz, 1H)
3.16 (d, J=12.8 Hz, 1H)
3.53 (d, d, J=12.8, 12.8 Hz, 1H)
3.59 (d, d, J=12.8, 12.8 Hz, 1H)
3.74 (d, J=12.8 Hz, 1H)
3.82 (d, J=12.8 Hz, 1H)
3.91 (d, d, d, d, J=12.8, 12.8, 12.8, 3.7 Hz, 1H)
6.71 (m, 3H)
m-H, p-H of Ph
o-H of Ph
H of AsPh3
H of AsPh3
6.97 (d, J=7.5 Hz, 2H)
7.27 (m, 6H)
7.35 (m, 9H)
magnitude of which increased at the expense of those of 1b
after successive addition of morpholine. New signals charac-
teristic of a ligated morpholine
Figure 1. Complex [PdI(Ph)
structure. Bottom: See Table 2 for the H NMR characterization.
A
were also observed (Table 2). Table 3. Crystal data and structure refinement for [PdI(Ph)
ACHTREUNG
Morpholine (5 equiv) was re-
quired to consume the 1b com-
pletely. The complex 1b was in
part restored after further addi-
tion of AsPh3 (1 equiv). Conse-
quently, morpholine reacted
with 1b to generate a new com-
plex 4bm in a reversible reac-
tion.
empirical formula
formula weight
T [K]293(2) K
l [Š]0.71069
crystal system, space group triclinic, P1
a [Š]10.378(11)
b [Š]11.464(7)
c [Š]12.498(10)
a [8]87.982(6)
703.74
index range
q range [8]2.26–32.00
ꢀ15ꢁhꢁ15
ꢀ17ꢁkꢁ17
ꢀ18ꢁlꢁ18
ACHTREUNG
¯
collected/unique reflections
27797/8838 [
q=32.00 [%]98.6
R
completeness to
refinement method
data/restraints/parameters
goodness-of-fit on
final
full-matrix least-squares on
8838/0/299
b [8]75.277(7)
g [8]64.197(4)
F2
1.062
R indices [I>2s(I)]
R1=0.0568, wR2=0.1749
R1=0.0787, wR2=0.1892
V [Š3]1289.86(16)
Z/1calcd [Mgmꢀ3]2/1.812
m [mmꢀ1]3.210
R indices (all data)
extinction coefficient
largest diff. peak/hole [eŠ
688
When morpholine (10 mmol)
was added to an NMR tube
0.0071(10)
ꢀ3]2.487/
ꢀ4.486
containing the dimer [PhPd
A
F(000)
G
I)(AsPh3)]2 (7b; 5 mmol) in
ACHTREUNG
668
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676

Sonagashira Reactions
FULL PAPER
Table 4. Relevant bond lengths [Š]and angles
[
8]for [PdI(Ph)-
(0.75 mL) using the integration of the doublet at d=7.1 ppm
(two o-H of Ph of 7b) and the integration of the two triplets
(one proton p-H and two protons m-H of Ph of 1b): KD =
0.013 (CDCl3, 258C). The value of K’=333mꢀ1/2 (CDCl3,
A
ꢀ
ꢀ
Pd1 C5
1.999(6)
2.128(5)
C10 C9
1.394(9)
1.399(8)
1.385(9)
1.403(9)
1.524(9)
1.360(12)
1.368(12)
1.511(10)
ꢀ
ꢀ
Pd1 N1
C10 C5
ꢀ
ꢀ
Pd1 As1
2.3621(11)
2.6728(15)
1.388(11)
1.437(13)
1.484(9)
C6 C5
1
258C) was determined by H NMR spectroscopy performed
ꢀ
ꢀ
Pd1 I1
C6 C7
on a solution of the dimer 7b (5 mmol) in CDCl3 (0.75 mL)
after addition of various amounts of morpholine (in the
range 0.5–3 equiv), by using the integration of the doublet
of the two o-H of Ph of 7b at d=7.1 ppm and that of the
doublet of the two o-H of Ph of 4bm at d=6.95 ppm (see
Figure 2 for the determination of K’). The value of K=
KDK’=4.3 (CDCl3, 258C) was thus deduced.
ꢀ
ꢀ
O1 C3
C4 C3
ꢀ
ꢀ
O1 C2
C8 C7
ꢀ
ꢀ
N1 C4
C8 C9
ꢀ
ꢀ
N1 C1
1.492(8)
C1 C2
C5-Pd1-N1
C5-Pd1-As1
N1-Pd1-As1
C5-Pd1-I1
N1-Pd1-I1
As1-Pd1-I1
C3-O1-C2
C4-N1-C1
C4-N1-Pd1
C1-N1-Pd1
93.1(2)
87.74(15)
178.75(16)
177.53(17)
87.92(16)
91.28(3)
109.8(6)
107.8(6)
120.6(4)
112.5(4)
C9-C10-C5
C5-C6-C7
N1-C4-C3
O1-C3-C4
N1-C1-C2
C6-C5-C10
C6-C5-Pd1
121.0(7)
120.7(6)
112.1(6)
112.4(7)
112.3(6)
117.7(6)
121.2(4)
121.1(5)
110.7(7)
C10-C5-Pd1
O1-C2-C1
The complex 4bm was also characterized by NMR spec-
troscopy (1H, 13C, and 2D), which revealed that the com-
plexation of the morpholine on the PdII center resulted in a
split of its eight protons which were then magnetically non-
ACHTREUNG
Figure 2. Determination of the equilibrium constant K’ for the reaction
between complex 7b and 4bm formed after addition of morpholine
(n equiv) to complex 7b (C0 =6.67 mm) in CDCl3 at 258C. K’’=
AHCTREUNG
[4bm]2/
([R2NH]2
[7b])=4x2/{C0
G
N
(1b) to generate complexes 4b. This reaction proceeds via
the dimeric complex 7b (Scheme 3).
4bm). Plot of x2 versus (1ꢀx)(nꢀ2x)2. The slope of the straight line gave
C
K’’=110900mꢀ1. K’=K’’1/2 =333mꢀ1/2 (Scheme 3).
Similar reactions were performed with piperidine. When
piperidine (10 mmol) was added to the dimer 7b (5 mmol) in
CDCl3 (0.75 mL), the signals of the dimer 7b totally disap-
peared. Two new complexes 4bp and 4’bp (Figure 3,
Scheme 3.
The equilibrium constant K (Scheme 3) was determined
from the value of KD and K’, with K=KDK’ [Eqs. (5)–(7)].
Figure 3. Proposed structures for 4bp and 4’bp.
Scheme 3) were characterized by 1H, 13C, and 2D NMR
spectroscopy. The structure of the major complex (63%)
was assigned to 4bp (Figure 3) with the piperidine and the
AsPh3 in a trans position, by analogy to the structure of
complex 4bm (Figure 1, top). The structure of isomeric
minor complex 4’bp (37%) could not be determined,
through lack of crystals. However, the signals of the o-H of
the minor complex are located at the lowest field (Table 5)
in comparison with those of complexes 4bm, 1b, or 7b, in
which the Ph group is in a position cis to AsPh3. This sug-
gests that the Ph group and AsPh3 are in a trans position in
½AsPh3Š½4 bŠ
ð5Þ
ð6Þ
ð7Þ
K ¼
½trans-1 bŠ½R2NHŠ
1=2
½AsPh3Š½7 bŠ
KD
¼
½trans-1 bŠ
½4 bŠ
K0 ¼
1=2
½R2NHŠ½7 bŠ
1
The value of KD was determined by H NMR spectrosco-
py from a solution (13.3 mm) of 1b (10 mmol) in CDCl3
Chem. Eur. J. 2007, 13, 666 – 676
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
669

A. Jutand et al.
Table 5. 1H NMR spectra (400 MHz, CDCl3, 258C, TMS) of 4bp and
4’bp (see Figure 3).
Therefore, one ligand AsPh3 can be substituted by a sec-
ondary amine R2NH (piperidine, morpholine, isopropyl-
Proton
d [ppm]
AHCTREUNG
4bp
A
ACHTREUNG
H4, H3, H5
1.2–1.7 (m, 6H)
AHCTREUNG
H2
H1
2.52 (dddd, J=12.8, 12.8, 12.8, 2.7 Hz, 2H)
3.34 (d, J=12.8 Hz, 2H)
3.49 (t, J=12.4 Hz, 1H)
6.73 (t, J=6 Hz, 1H)
G
ACHTREUNG
values of the equilibrium constants in Table 6, we deduce
that AsPh3 is much more labile than PPh3, because the
ꢀ
N H
p-H
m-H
o-H
H of AsPh3
4’bp
H4, H3, H5
6.75 (t, J=6 Hz, 2H)
7.01 (d, J=6 Hz, 2H)
7.3–7.4 (m, 15H)
dimer [PhPd
[PdI(Ph)
(AsPh3)2],[5] whereas [PhPd
known and synthesized independently,[10] has never been ob-
served to be formed from trans-[PdI(Ph)(PPh3)2].
(m-I)
ACHTREUNG
U
R
ACHTREUNG
AHCTREUNG
1.2–1.7 (m, 6H)
H2
2.60 (pseudo q, 2H)
2.95 (t, J=12.4 Hz, 1H)
3.24 (d, J=13 Hz, 2H)
7.05 (t, J=7 Hz, 1H)
7.09 (t, J=7 Hz, 2H)
7.27 (m, 2H)
ꢀ
N H
H1
m-H and p-H
m-H and p-H
o-H
H of AsPh3
7.3–7.4 (m, 15H)
complex 4’bp. Therefore, from the two possible isomers, we
propose the structure of 4’bp in Figure 3.
Whatever the ligand, the ability of the amine to substitute
one AsPh3 or one PPh3 decreases in the order: piperidine>
morpholine>tert-butylamine>diisopropylamine.
The complex trans-1b reacted with piperidine to give
complexes 4bp and 4’bp in a reversible reaction (Scheme 3).
Two equivalents of piperidine were required to shift the
equilibrium completely toward complexes 4bp and 4’bp. A
Piperidine is more basic than morpholine (Table 6)[11] and
is thus a better ligand for the PdII center. Although diisopro-
pylamine is more basic than morpholine, it is a poorer
ligand for the PdII center. Thus the AsPh3 substitution is
also sensitive to the steric hindrance of the amine.
minimum value of K’=[4bp+4’bp]/
([7b]1/2[piperidine])>
A
850mꢀ1/2 (CDCl3, 258C) could be estimated from the reac-
tion of dimer 7b with piperidine (1 equiv) by using the
1H NMR data. Since KD =0.013 (vide supra), a minimum
value of K=KDK’>11 (CDCl3, 258C) could be estimated
(Table 6).
Role of the amine in the oxidative addition: Complexes
trans-[PdX(Ar)(L)2](L =PPh3 or AsPh3) are formed in the
first step of the catalytic cycle of the Sonogashira reaction
by the oxidative addition of ArX with [Pd0(L)2]complexes.
We have now established that secondary amines can coordi-
nate trans-[PdX(Ar)(L)2]complexes with the release of one
ligand L, which can in turn react with [Pd0(L)2]to form un-
reactive [Pd0(L)3]complexes in the course of the oxidative
addition. A decelerating effect of the amine in the rate of
the overall oxidative addition might therefore be expected
(path a in Scheme 4). It was thus of great interest to probe
the effect of amine on the kinetics of the oxidative addition.
Table 6. Equilibrium constants K for the substitution of one ligand L by
an amine in complexes trans-[PdI(Ph)(L)2] (1) in chloroform at 258C
[Eq. (8)].
Amine (pKa)[a]
K
L=PPh3[3]L
=AsPh3
>11
piperidine (11.12)
0.11
morpholine (8.33)
tert-butylamine (10.83)
diisopropylamine (10.96)
0.014
0.002
0.00007
4.3
0.041
0.007
[a]In water (ref. [11]).
The complex 4bd was formed by addition of a large
excess of diisopropylamine to complex trans-1b in CDCl3
(Scheme 3). Only after addition of 30 equivalents of diiso-
propylamine was complex 4bd formed in a significant
amount (46%). The complex was detected through the pres-
ence of the protons of the Ph group ligated to the PdII
center. The ligated iPr2NH could not be characterized be-
cause its signals overlapped those of the free iPr2NH added
in a large excess. The equilibrium constant K (Scheme 3) is
given in Table 6. In addition, one AsPh3 from complex
trans-1b was substituted by a primary amine such as tert-bu-
tylamine to give the complex [PdI(Ph)
N
N
(4bt) in a reversible reaction (see the K value in Table 6).
Scheme 4.
670
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676

Sonagashira Reactions
FULL PAPER
d½Pd0Š
dt
ðk3 þ k4K2½R2NHŠ=½LŠÞ½PhIŠ
Moreover, the coordination of Pd0 complexes by amines was
recently under debate.[12]
ð10Þ
¼ ꢀ
½LŠ=K1 þ K2½R2NHŠ=½LŠ
k3 þ k4K2½R2NHŠ=½LŠ
A
ACHTREUNG
ACHTREUNG
ð11Þ
kapp
¼
½LŠ=K1 þ K2½R2NHŠ=½LŠ
AHCTREUNG
At low [R2NH], K2
A
AHCTREUNG
fies to Equation (12); under stoichiometric conditions
portional to its concentration) measured at a rotating gold
([PhI]=[Pd0L4]=C0), so Equation (13) is obtained, in which
disk electrode polarized at +0.2 V versus SCE on the pla-
x=molar fraction of [Pd0
ACHTREUNG
teau of the oxidation wave of [Pd0
(PPh3)3], first in the ab-
T
rent of [Pd0
[Pd0
ACHTREUNG
sence of amine and then in the presence of various amounts
of piperidine or morpholine, added to the Pd0 complex
before PhI. The graph of the molar fraction x of the Pd0
complex versus time shows that the oxidative addition was
faster when performed in the presence of amine (piperidine
or morpholine) (Figure 4 left). A 31P NMR measurement at
the end of the oxidative addition performed in the presence
of piperidine (100 mm) did exhibit the singlet of trans-
AHCTREUNG
k3K1 k4K1K2½R2NHŠ
ð12Þ
ð13Þ
kapp
¼
þ
2
½LŠ
½LŠ
1
x
¼ kappC0t þ 1
The plots of 1/x versus time were linear, in agreement
[PdI(Ph)(PPh3)(piperidine)]at d=32.5 ppm, as already char-
H
acterized.[3] This proves that the oxidative addition worked
well in the presence of the amine.[13] Although PPh3 was re-
leased in the course of the oxidative addition performed in
the presence of piperidine, the oxidative addition went
with Equation (13) (Figure 4 middle). The values of kapp
were determined from the slope of the straight lines. The
plot of kapp versus piperidine concentration was linear
(Figure 4 right) in agreement with Equation (12). From the
intercept, one obtains k3K1 =0.11 sꢀ1 (DMF, 208C), which
faster. This indicates that the reactive species differed from
characterizes the reactivity of [Pd0
(PPh3)4]alone (this value
A
0
[Pd0
(PPh3)
A
is similar to that determined in previous work).[7] From the
slope of the straight line, k4K1K2 =0.013 sꢀ1 (DMF, 208C)
was determined.[15]
G
(amine)](path b in Scheme 4). Consequently, the de-
celerating effect due the release of PPh3 by the amine (path
a in Scheme 4) and the more effective accelerating effect
due to the formation of a more reactive species [Pd0
G
Role of the amine in the step involving the alkyne: It is now
A
well established that trans-[PdI(Ph)(L)2]complexes (L
PPh3 or AsPh3) react with a secondary amine (morpholine
or piperidine) to generate [PdI(Ph)L(amine)]complexes in
reversible reactions. Consequently, the amine may compete
with the alkyne for the complexation of the PdII center in
trans-[PdI(Ph)(L)2]complexes.
=
The accelerating effect was less important in the presence
of morpholine than with piperidine added at the same con-
centration (Figure 4 left), suggesting that morpholine is a
less good ligand for [Pd0(L)]than piperidine. The accelerat-
ing effect was more pronounced when the concentration of
the piperidine was increased (Figure 4 middle). The kinetic
law for the mechanism proposed in Scheme 4 (paths a and b
operating in parallel) is given in Equations (9)–(11).
ACHTREUNG
Experiments were undertaken to characterize such com-
petition, starting from trans-[PdI(Ph)
A
mainly from the dimer [PhPd(m-I)(AsPh3)]2 (7b), since the
A
ACHTREUNG
complexation of the amine proceeds via 7b (vide supra).
Moreover, the complexation of the alkyne must also go by
rate ¼ k3½PhIŠ½Pd0ðLÞ2Š þ k4½PhIŠ½Pd0ðLÞðR2NHފ
ð9Þ
ꢂ
the same route, via 7b. It is worth noting that alkynes (HC
Figure 4. Kinetics of the oxidative addition of PhI (2 mm) to [Pd0
ACHTREUNG
the Pd0 complex versus time (x=i/i0; i=oxidation current of [Pd0
A
ACHTREUNG
^
*
(
&
*
dine [(”) 10 mm, ( ) 100 mm, ( ) 150 mm]. Right: kapp versus piperidine concentration [see Eq. (12)].
Chem. Eur. J. 2007, 13, 666 – 676
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
671

A. Jutand et al.
ꢀ
C R) may react with trans-[PdI(Ph)(L)2]complexes in the
We have seen above that the reaction of the dimer 7b
with a stoichiometric amount of piperidine and alkyne 5
(7b/piperidine/5=1:2:2) did not deliver the coupling prod-
uct. However, when the amount of piperidine was doubled
ꢀ
absence of base in a carbopalladation step, leading to [Ph
[16]
ꢀ
CH=CH(R) PdI(L)2]complexes,
or even undergo multi-
carbopalladation steps.[16c] Carbopalladation is favored with
electron-deficient alkynes (for example, R=CO2Et).[16] Our
present investigation mainly concerns the electron-rich
alkyne 5, which has been used in catalytic Sonogashira reac-
tions (Scheme 1). Alkyne 5 did not react with piperidine or
morpholine, as monitored by 1H NMR spectroscopy in
CDCl3.
ꢀ ꢂ ꢀ
(7b/piperidine/5=1:4:2), the coupling product Ph C C
ꢀ
ꢀ
CH2 CH2 OH (6) was formed in approximately 50% yield,
1
as determined by H NMR spectroscopy with CH2Cl2 as an
internal standard (Figure 5). Just after mixing, the free
The alkyne 5 was added to an NMR tube containing the
dimer 7b (0.5 mmol) in CDCl3 (0.75 mL; 7b/5=1:2). The
protons of the Ph group attached to the PdII center re-
mained unchanged, but broad signals were observed for the
ꢀ
CH2 CH2 protons of 5. The resulting complex could not be
characterized any better. Under similar conditions, stoichio-
metric amounts of piperidine and 5 (7b/piperidine/5=1:2:2)
were introduced to an NMR tube containing 7b. At short
1
times the H NMR signals of complexes 4bp and 4’bp and
*
Figure 5. Concentration profiles of alkyne 5 ( ) and coupling product 6
of the free 5 were exhibited. The competition between 5
and piperidine (used at the same concentration) for the co-
ordination of the PdII center is thus in favor of the piperi-
dine. Such an effect will be amplified if the amine is used as
solvent, as in the present catalytic reactions [Eq. (3)]. The
*
(
) in the reaction of dimer 7b with piperidine and alkyne 5 (7b/piperi-
dine/5=1:4:2) in CDCl3 at 258C.
alkyne 5 was observed together with complexes 4bp and
4’bp and free piperidine in an almost stoichiometric
amount: piperidine/(4bp+4’bp)=1.08:1, in agreement with
the results established above: that is, 4bp and 4’bp were
formed quantitatively from 7b as soon as 7b/piperidine=
1:2. This again shows that piperidine is more reactive than 5
in the cleavage of the iodide bridges of 7b to generate com-
plexes 4bp and 4’bp. The coupling product 6 then appeared
over time (Figure 5), but its formation was not quantitative.
The complexes 4bp and 4’bp, always in the same ratio as in-
itially, were still detected (57%) at the end of the reaction,
whereas the starting alkyne 5 was no longer observed
(Figure 5).
ꢀ
ꢀ
stability order, with R=CH2 CH2 OH could thus be estab-
lished
(AsPh3)
as:[17]
[PdI(Ph)(AsPh3)(piperidine)]>[PdI(Ph)-
G
2
ꢂ
A
(h -RC CH)]. At longer times, a more complex
1H NMR spectrum was obtained as in the reaction of isolat-
ed 4bp and 4’bp with 5 (vide infra) but the coupling product
6 was not formed.
ꢂ
PhC CPh was selected for another experiment, reaction
ꢂ
of the dimer 7b (6.67 mm in CDCl3) with PhC CPh (7b/
ꢂ
PhC CPh=1:2), because it cannot react in a Sonogashira
reaction and moreover it is supposed to undergo a slow car-
bopalladation after coordination of the PdII atom of a Ph–
PdII complex. The first H NMR spectrum recorded exhibit-
1
ed the signals of 7b and those of a new complex 8b. After
1 h30 min, only 8b was observed [Eq. (14)], characterized
by 1H NMR spectroscopy and FAB+ MS. After further addi-
tion of piperidine (1 equiv) to complex 8b, new signals ap-
peared together with some free piperidine (56%). The com-
plex 4bp, which might have been formed by substitution of
From the concentration profiles of 5 and 6 shown in
Figure 5 and the information obtained in Equations (14) and
(15), a mechanism is proposed in Equations (16)–(20), in
which R2NH=piperidine, for the formation of the coupling
product 6. To simplify the discussion, the proposed mecha-
nism involves only complex 4bp. A similar mechanism may
involve the isomer 4’bp as well. After the fast formation of
complex 4bp, the ligand AsPh3 is substituted by the alkyne
5 (R’=CH2CH2OH) to give complex 10p [Eq. (17)]. The li-
gated alkyne is more acidic than the free alkyne and can be
deprotonated by an extra piperidine [Eq. (18)]to complex
11p, which gives the coupling product 6 by reductive elimi-
nation [Eq. (19)].
ꢂ
PhC CPh by piperidine in complex 8b, was not present;
ꢂ
this fact proves that PhC CPh is a better ligand than piperi-
dine for the PdII in complex 8b. Signals characteristic of a li-
gated piperidine were present together with free AsPh3.
This suggests the formation of complex 9p [Eq. (15)].
672
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676

Sonagashira Reactions
FULL PAPER
quently, a new complex was formed which also contained a
ligated piperidine. The detection of free AsPh3, which
turned to arsine oxide with time, was interesting. Even if
complex 10p could not then be fully characterized, because
1
its H NMR spectrum is too complicated due to the protons
of three different ligated piperidines, the release of AsPh3 is
nevertheless indicative of the reaction given in Equa-
tion (17).
Mechanism of the copper-free palladium-catalyzed Sonoga-
shira reaction: Having to hand the mechanism of Equa-
tions (16)–(20) established for aryl–PdII complexes ligated
by AsPh3, we can rationalize the results of the catalytic So-
nogashira reactions reported in Table 1. The fact that the
Early in the reaction (t<60 min, Figure 5), the rates of
formation of 6 and disappearance of 5 follow a similar ki-
netic law. This means that the reductive elimination
[Eq. (19)]is faster than the rate-determining reaction con-
sisting of the equilibrium in Equation (17) shifted by the re-
action in Equation (18). After 80 min, the decrease in 5 is
more pronounced than during the first part of the reaction,
and the formation of 6 nearly stops. The fact that 4bp and
4’bp were recovered in about 56% yield while 5 was totally
consumed suggests that 5 was involved in a reaction compet-
ing with the reaction given in Equation (17). The alkyne 5
must be involved in the coordination of the Pd0 complex
formed in the reductive elimination [Eq. (20)].[18] This is
why the reaction of 4bp and 4’bp with 5 was not quantita-
tive. Such a competitive reac-
catalytic reactions involving the same precursor [Pd0
(dba)2]
G
were more efficient when the ligand PPh3 was used instead
of AsPh3, although AsPh3 was more easily substituted than
PPh3 by an amine in trans-[PdI(Ph)(L)2]complexes, indi-
cates that two different catalytic cycles can operate, accord-
ing to the ligand and the amine. The two mechanisms (paths
A and B in Scheme 5) are branched at the level of trans-
[PdI(Ph)(L)2]complexes generated in the fast oxidative ad-
dition step.
When L=PPh3, its substitution by the amine in trans-
[PdI(Ph)
pholine and piperidine in Table 6) because of the low con-
centration of the dimer [{Pd(m-I)(Ph)(PPh3)}2]; consequently
the path A is dominant. According to path A, we under-
A
A
ACHTREUNG
tion will be less efficient in cat-
alytic reactions since the Pd0
complex will always be stabi-
lized by the two AsPh3 intro-
duced with the Pd0 precursor.
The feasibility of the reaction
given in Equation (17) was
tested by reaction of a mixture
of isolated 4bp and 4’bp
(62:38) with a stoichiometric
amount of alkyne 5. The reac-
tion
was
monitored
by
1H NMR spectroscopy. A very
complicated spectrum was ob-
tained in which we could distin-
guish some unreacted 4bp and
4’bp by their respective ligated
piperidine groups (Table 5).
The signals of free alkyne 5 dis-
appeared slowly. New signals
assigned to a ligated piperidine
(d1 =2.98 ppm (d, J=14 Hz,
1H), d2 =3.16 ppm (t, J=
12.5 Hz, 2H)) could be distin-
guished from those of the ligat-
ed piperidine of 4bp and 4’bp,
located at higher field as in
ꢀ
ꢀ
Scheme 5. Mechanism of the copper-free palladium-catalyzed Sonogashira reaction (R=CH2 CH2 OH). Path
A: the amine is a less good ligand than the alkyne for the PdII center in [PdX(Ar)(L)2](L =PPh3 with amine=
piperidine or morpholine). Path B: the amine is a better ligand than the alkyne for the PdII center in
complex 9p [Eq. (15)]. Conse- [PdX(Ar)(L)2](L =AsPh3, amine=piperidine).
Chem. Eur. J. 2007, 13, 666 – 676
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
673

A. Jutand et al.
stand why [Pd0
E
N
A
(m-I)
ACHTREUNG
2PPh3: the excess of free ligand in the former case inhibits
the substitution of PPh3 by the alkyne. In the case of PPh3,
piperidine was more efficient than morpholine (entries 1
and 3 in Table 1), because it is more basic and consequently
more reactive in the deprotonation step.[19]
ꢀ ꢂ ꢀ
C
C
ꢀ
ꢀ
CH2 CH2 OH (6) [Eq. (3)]: The procedure was similar to that reported
[1a,6]
for [Pd0
G
by using the same amount and concentra-
(dba)2]
tion of the pre-catalyst and reagents to allow comparison. [Pd
ACHTREUNG
(28.7 mg, 0.05 mmol) was introduced under an argon atmosphere in pi-
peridine (3 mL), followed by AsPh3 (31 mg, 0.1 mmol). PhI (110 mL,
When L=AsPh3, its substitution in trans-[PdI(Ph)-
ꢂ ꢀ
ꢀ
ꢀ
1 mmol) was then added, followed by HC C CH2 CH2 OH (5)
(152 mL, 2 mmol). The solution was stirred at room temperature for
3 h15 min. It was then hydrolyzed with aqueous NH4Cl, extracted with
diethyl ether, and dried on MgSO4. After evaporation of the solvent,
CHCl2CHCl2 (75 mL) was added as internal standard. The yield of 6 and
the unreacted PhI was determined from the 1H NMR spectrum of the
crude product after calibration (Table 1). Similar experiments were per-
formed using morpholine (3.5 mL) instead of piperidine and PPh3
(26.2 mg, 0.1 mmol) instead of AsPh3 during the same reaction time
ACHTREUNG
A
ACHTREUNG
by the alkyne 5 (present work, vide supra). Consequently,
the mechanism of path B established in this work is domi-
nant. In presence of piperidine, yields were lower when
AsPh3 was used instead of PPh3 (entries 1 and 2 in Table 1).
This means that path B is less efficient than path A. In the
case of AsPh3, piperidine appears to be slightly less efficient
than morpholine (entries 2 and 4 in Table 1). This may be
rationalized by the fact that the substitution of one AsPh3
1
3
ACHTREUNG
(Table 1). H NMR (250 MHz, CDCl3, 258C, TMS): d=2.03 (t, J
(H,H)=
2.5 Hz, 1H), 2.69 (t, 3J(H,H)=6.3 Hz, 2H), 3.81 (t, 3J
ACHTREUNG
2H), 7.26–7.30 (m, 3H), 7.40–7.43 ppm (m, 2H), similar to an authentic
sample.[1a,6]
by morpholine in trans-[PdI(Ph)(AsPh3)2]is less easy than
R
Generalprocedure for the reaction of amines with the dimer [PhPd
ACHTREUNG
the substitution by piperidine. The mechanism in path A
may thus operate partially when morpholine is used.[17] It is
worth noting that one may easily pass from path A to path
B, and vice versa, by changing the relative concentrations of
the alkyne and amine.
1
(AsPh3)]2
ACHTREUNG
of amines: piperidine (2 equiv) or morpholine (1, 2, and 3 equiv) or diiso-
propylamine (2, 4, 10, and 60 equiv) (n=total number of equivalents
added) were added to dimer 7b (6.1 mg, 5 mmol) in CDCl3 (0.75 mL).
The 1H NMR spectrum was recorded after each addition of amine to
detect the formation of [PhPd
Generalprocedure for the reaction of amines with trans-[PdI(Ph)-
(AsPh3)2] (1b), as monitored by 1H NMR spectroscopy: The procedure
(AsPh3)2
(amine)]( 4b).
AHCTREUNG
Conclusion
was the same as above, starting from a solution of 1b (9.2 mg, 10 mmol)
in CDCl3 (0.75 mL) to which were added various amounts of amines: pi-
peridine (n=1, 2 and 3 equiv) or morpholine (0.5, 1, 1.5, 2, 3, and
6 equiv) or diisopropylamine (n=1, 6 and 30 equiv) or tert-butylamine
(n=1, 2, and 5 equiv) (n=total number of equivalents added).
Amines used as bases in copper-free palladium-catalyzed
Sonogashira reactions play a very important and multiple
roles. Besides their expected function as deprotonating
agents, amines (morpholine, piperidine) may be involved in
two different steps preceding the deprotonation: 1) they in-
terfere in the oxidative addition by an accelerating effect
Synthesis of [PdI(Ph)
ACHTREUNG
pholine with [PhPd(m-I)
N
G
ACHTREUNG
was added to a solution of dimer 7b (6.1 mg, 5 mmol) in chloroform
(0.75 mL). Single crystals of 4bm were obtained by diffusion of pentane
into the chloroform solution. 1H NMR (400 MHz, CDCl3): see Table 2;
13C NMR (63.3 MHz, CDCl3, 258C, TMS): d=48.65, 53.81, 127.98,
128.40, 129.78, 133.45, 133.93, 134.76 ppm; X-ray structure: see Figure 1
(top) and Tables 3 and 4.
due to the formation of more reactive [Pd0L
(amine)]com-
U
plexes; and 2) amines can substitute one ligand in trans-
[PdI(Ph)(L)2]complexes formed in the oxidative addition
step, the substitution of AsPh3 being more favored than that
of PPh3. Depending on the rate of the competition between
amine and alkyne in the substitution of one L in trans-
[PdI(Ph)(L)2], two different mechanisms may operate (path
A or B in Scheme 5). Consequently, the amine does not
react as a simple base in Sonogashira reactions, but it is also
involved as a ligand of aryl–PdII complexes.
Synthesis of [PdI(Ph)
of piperidine with [PhPd
0.4 mmol) were added to
ACHTREUNG
N
A
ACHTREUNG
a
0.1 mmol) in chloroform (10 mL). After 30 min, the solution was filtered
and the solvent eliminated under vacuum. Yellow-green crystals were
formed. Two successive crystallizations from chloroform/pentane gave
yellow crystals (93 mg, 67% yield), as a mixture of complexes 4bp and
4’bp that could not be separated. 1H NMR (400 MHz, CDCl3, 258C,
TMS): see Table 5. 13C NMR (63.3 MHz, CDCl3, 258C, TMS): d=23.59
(23.79), 27.68 (27.79), 50.01 (52.06), 127.68, 128.32, 128.58, 129.63, 133.66,
133.97, 134.23, 134.98 ppm; elemental analysis calcd (%) for
C29H31AsINPd (700.98): C 49.63, H 4.45, N 2.00; found C 49.29, H 4.51,
N 1.74.
ExperimentalSection
A
ACHTREUNG
AHCTREUNG
General: 31P NMR spectra were recorded on a Bruker spectrometer
(101 MHz) in DMF containing 10% of [D6]acetone. 1H NMR spectra
were recorded on a Bruker spectrometer (250 or 400 MHz). Cyclic vol-
tammetry and amperometry were performed at gold disk electrodes with
a home-made potentiostat and a Tacussel GSTP4 waveform generator.
The voltammograms were recorded on a Nicolet 301 oscilloscope.
(42 mL, 0.3 mmol) in CDCl3 (0.75 mL). H NMR (250 MHz, CDCl3, 258C,
1
3
ꢀ
TMS): d=6.66 (m, 3H; m-H and p-H of Ph Pd), 7.00 (d, J
A
ꢀ
2H; o-H of Ph Pd), 7.2–7.4 ppm (m; H of AsPh3). The ligated iPr2NH
could not be characterized due to the overlapping of its signals with
those of the free iPr2NH added in excess. The complex 4bd was not iso-
lated because it formed a mixture (46:54) with the starting complex 1b.
Chemicals: DMF was distilled from calcium hydride under vacuum and
A
ACHTREUNG
kept under argon. CDCl3, PhI, AsPh3, morpholine, piperidine, diisopro-
a
ACHTREUNG
ꢂ ꢀ
ꢀ
ꢀ
ꢂ
pylamine, tert-butylamine, HC C CH2 CH2 OH, and PhC CPh were
(9.2 mg, 10 mmol) in CDCl3 (0.75 mL). The complex 4bt was not isolated
because it formed a mixture (2.2:10) with the starting complex 1b, but it
commercial products. [Pd
(dba)2],[20] [Pd(PPh3)4],[21] trans-[PdI(Ph)-
A
ACHTREUNG
674
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676

Sonagashira Reactions
FULL PAPER
was characterized in situ. 1H NMR (250 MHz, CDCl3, 258C, TMS): d=
[1]a) M. Alami, F. Ferri, G. Linstrumelle, Tetrahedron Lett. 1993, 34,
6403–6406; b) J. F. Nguefack, V. Bolitt, D. Sinou, Tetrahedron Lett.
1996, 37, 5527–5530; c) V. P. W. Bçhm, W. A. Herrmann, Eur. J.
Org. Chem. 2000, 3679–3681; d) R. G. Heidenreich, K. Kçlher,
J. G. E. Krauter, J. Pietsch, Synlett 2002, 1118–1122; e) M. Pal, K.
Parasuraman, S. Gupta, K. R. Yeleswarapu, Synlett 2002, 1976–
1982; f) T. Fukuyama, M. Shinmen, S. Nishitani, M. Sato, I. Ryu,
Org. Lett. 2002, 4, 1691–1694; g) N. E. Leadbeater, B. J. Tominack,
Tetrahedron Lett. 2003, 44, 8653–8656; h) A. Soheili, J. Albaneze-
Walker, J. A. Murry, P. G. Dormer, D. L. Hughes, Org. Lett. 2003, 5,
4191–4194.
[2]For mechanisms in which the pronucleophile is first ligated to the
PdII center of [PdX(Ar)Ln]complexes before its deprotonation by
an extra base, see: a) J. P. Wolfe, S. Wagaw, J. F. Marcoux, S. L.
Buchwald, Acc. Chem. Res. 1998, 31, 805–818; b) J. F. Hartwig,
Angew. Chem. 1998, 110, 2154–2177; Angew. Chem. Int. Ed. 1998,
37, 2047–2067; c) X. Moreau, J. L. Campagne, G. Meyer, A. Jutand,
Eur. J. Org. Chem. 2005, 3749–3760.
ꢀ
1.18 (s, 9H; tBu), 6.66 (m, 3H; m-H and p-H of Ph Pd), 6.92 (m, 2H; o-
ꢀ
H of Ph Pd), 7.2–7.4 ppm (m; H of AsPh3).
Determination of the equilibrium constant K’ between 7b and 4bp (see
main text and Scheme 3): Various amounts of morpholine (0.5–3 equiv)
were added to a solution of dimer 7b (6.1 mg, 5 mmol, 6.67 mm) in CDCl3
(0.75 mL). The 1H NMR (250 MHz) spectrum was recorded after each
addition of morpholine. See Figure 2 for the calculation of K’.
2
ꢂ
Formation of [PdI(Ph)(h -PhC CPh)
A
ꢂ
CPh with the dimer 7b: PhC CPh (1.8 mg, 10 mmol) was added to the
dimer 7b (6.1 mg, 5 mmol) in CDCl3 (0.75 mL). The 1H NMR spectrum
was recorded with time. The first spectrum revealed a mixture of 7b and
8b. After 1 h30 min, only 8b was observed. 1H NMR (250 MHz, CDCl3,
ꢀ
ꢀ
258C, TMS): d=6.50 (m, 1H; p-H of Ph Pd), 6.59 (m, 2H; m-H of Ph
Pd), 6.87 (m, 6H; H of PhC=CPh), 6.97 (m, 4H; H of PhC=CPh), 7.19
(d, 3J
(H,H)=5 Hz, 2H; o-H of Ph Pd); 7.25 (m; H of AsPh3), 7.32 ppm
ꢀ
(m; H of AsPh3); FAB+ MS: m/z: 667 [M+ꢀI], 591 [M+ꢀIꢀPh].
2
ꢂ
Formation of [PdI(Ph)(h -PhC CPh)(piperidine)] (9p) by reaction of
piperidine with complex (8b): Piperidine (1 mL, 10 mmol) was added to
an NMR tube containing 8b formed as described above. The reaction
[3]A. Jutand, S. Negri, A. Principaud, Eur. J. Inorg. Chem. 2005, 631–
635.
[4]a) V. Farina, B. J. Krisnan, J. Am. Chem. Soc. 1991, 113, 9585–9595;
b) “Recent advances in the Stille reaction”: V. Farina, G. P. Roth,
Adv. Met.-Org. Chem. 1996, 5, 1–53; c) A. L. Casado, P. Espinet, J.
Am. Chem. Soc. 1998, 120, 8978–8985.
was followed by 1H NMR spectroscopy. 1H NMR (250 MHz, CDCl3,
3
258C, TMS): d=1.0–1.5 (m, 4H), 2.12 (t, J
(H,H)=10.8 Hz, 1H), 2.31 (d,
3J
A
ACHTREUNG
A
ACHTREUNG
2H), 6.72–7.11 (m, 13H), 7.23–7.40 ppm (m; H of AsPh3). Free AsPh3 at
[5]C. Amatore, A. A Bahsoun, A. Jutand, G. Meyer, A. Ndedi Ntepe,
L. Ricard, J. Am. Chem. Soc. 2003, 125, 4212–4222.
[6]F. Ferri, PhD Thesis, UniversitØ Paris VI, 1995, p. 63.
[7]C. Amatore, A. Jutand, F. Khalil, M. A. MꢁBarki, L. Mottier, Orga-
nometallics 1993, 12, 3168–3178.
[8]C. Amatore, A. Bucaille, A. Fuxa, A. Jutand, G. Meyer, A. Ndedi
Ntepe, Chem. Eur. J. 2001, 7, 2134–2142.
[9]For amine complexes formed by reaction of amines with dimeric
d=7.33 ppm was also observed as a singlet.
ꢀ ꢂ ꢀ
ꢀ
ꢀ
Kinetics of formation of the coupling product Ph
ꢂ ꢀ
C C
CH2 CH2 OH
ꢀ
ꢀ
(6) from 7b and HC
C CH2 CH2 OH (5) in the presence of piperi-
1
dine, as monitored by H NMR spectroscopy (Figure 5): Piperidine (2 mL,
20 mmol), followed by 5 (0.8 mL, 10 mmol), was added to the dimer 7b
(6.1 mg, 5 mmol) in CDCl3 (0.75 mL). A known amount of CH2Cl2 was
added as an internal standard. The reaction was monitored by 1H NMR
spectroscopy. The amounts of 6 and unreacted 5 were determined by
[(Pd
N
N
ꢂ ꢀ
comparing the integration of their respective protons C C CH2 versus
those of CH2Cl2 (Figure 5).
ganometallics 1995, 14, 3030–3039; b) R. A. Widenhoefer, A. H.
Zhong, S. L. Buchwald, Organometallics 1996, 15, 2745–2754, and
reference [2a].
Electrochemical procedure for cyclic voltammetry and for the kinetics of
the oxidative addition of PhI to [Pd0
(PPh3)4] in the presence of piperi-
R
[10]V. V. Grushin, Organometallics 2000, 19, 1888–1900.
dine or morpholine: Experiments were carried out in a three-electrode
thermostated cell (208C) connected to a Schlenk line. The reference was
a saturated calomel electrode (Radiometer) separated from the solution
by a bridge filled with DMF (3 mL) containing nBu4NBF4 (0.3m). The
counter electrode was a platinum wire, apparent surface area ꢃ1 cm2.
DMF (15 mL) containing nBu4NBF4 (0.3m) was introduced to the cell
[11]A basicity scale in aqueous solution is given in:
,
Handbook of
Chemistry and Physics, 52nd ed. (Ed.: R. C. Weast), The Chemical
Rubber Co., Ohio, 1971–1972, p. D-177.
[12]a) U. K. Singh, E. R. Strieter, D. G. Blackmond, S. L. Buchwald, J.
Am. Chem. Soc. 2002, 124, 14104–14114; b) S. Shekhar, P. Ryberg,
J. F. Hartwig, Org. Lett. 2006, 8, 851–854; c) S. Shekhar, P. Ryberg,
J. F. Hartwig, J. F. Matthew, D. G. Blackmond, E. R. Strieter, S. L.
Buchwald, J. Am. Chem. Soc. 2006, 128, 3584–3591.
followed by [Pd(PPh3)4](34.7 mg, 0.03 mmol). Cyclic voltammetry was
U
performed at a steady gold disk electrode (d=2 mm) at a scan rate of
0.2 Vsꢀ1 in the absence of amine and then in the presence of an increas-
ing amount (5–75 equiv) of amine (piperidine or morpholine).
[13]a) It is reported that [Pd 0 (L)n]complexes may react with amines
R2NH in oxidative addition leading to palladium(II) complexes
The kinetic measurements for the oxidative addition of PhI were per-
formed at a rotating gold disk electrode (Radiometer, EDI 65109, d
2 mm, angular velocity: 105 radsꢀ1) polarized at +0.2 V versus SCE. PhI
(3.4 mL, 0.03 mmol) was added and the decrease in the oxidation current
was recorded versus time until conversion was complete. Other experi-
ments were performed similarly in the presence of piperidine (148 mL,
1.5 mmol) or morpholine (131 mL, 1.5 mmol), added before the PhI.
ꢀ
ꢀ
[R2N Pd HLn’](L =PCy3) (references [13b,c]). When increasing
amounts of morpholine or piperidine were added to [Pd0
(PPh3)4], its
AHCTREUNG
oxidation peak did not disappear but was slightly shifted to more
positive potentials. It became broader while the oxidation peak cur-
rent increased. This indicates that the new complex(es) formed in
solution were still palladium(0) complexes involved in equilibrium
with the amine. Work is in progress to characterize such species
better. Consequently, no oxidative addition of the amine with the
Pd0 complex occurred during the timescale investigated here
(10 min between each addition of amine), which is much longer
than the time required for the oxidative addition (Figure 4 left);
b) T. Yamamoto, K. Sano, A. Yamamoto, Chem. Lett. 1982, 907–
910; c) A. I. Siriwardana, M. Kamada, I. Nakamura, Y. Yamamoto,
J. Org. Chem. 2005, 70, 5932–5937.
CCDC 605796 contains the supplementary crystallographic data for com-
plex 4bm. These data can be obtained free of charge from the Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Acknowledgements
[14]In the absence of excess PPh 3, the concentration of PPh3 varies from
C0 to 3C0 during the course of the oxidative addition due to its con-
tinuous release in the three equilibria (K1, K, and K2). Consequently,
the variation of the concentration of L should have been considered
in the integration of the kinetics law with [L]=(3ꢀ2x)C0. Due to
the complexity of the kinetics law, a constant value of [L]was con-
This work has been supported in part by the Centre National de la Re-
cherche Scientifique (UMR CNRS-ENS-UPMC 8640) and the Minist›re
de la Recherche (École Normale SupØrieure). Dr. Guillaume Prestat
(UniversitØ Pierre et Marie Curie, Paris VI) is thanked for a helpful dis-
cussion. The authors thank Johnson Matthey for loan of a palladium salt.
Chem. Eur. J. 2007, 13, 666 – 676
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
675

A. Jutand et al.
sidered for the integration of the kinetics law, equal to its average
value of 2C0 (C0 =2 mm).
[15]a) From the expression for kapp given in Equation (11), we should
merous signals, we could distinguish some free morpholine, which is
expected since more than two equivalents of morpholine are re-
quired to cleave all the iodide bridges. The free alkyne 5 was not ob-
served. It seems that coordination of both the morpholine and the
alkyne on the PdII center occurs when they are used at the same
concentration.
observe a limit value at high R2NH concentrations (K2
[L]/K1,) with app =k4, that is, when the equilibrium between
[Pd0(L)2]and [Pd 0(L)
(amine)]is totally shifted to the formation of
[Pd0(L)
(amine)], which would then be the unique reactive complex.
A
k
AHCTREUNG
0
U
[18]For complexation of Pd alkynes, see: a) C. Amatore, S. Bensalem,
However, the reactions went too fast at high piperidine concentra-
tions (for example, t1/2 =3.5 s for [piperidine]=150 mm), preventing
accurate data at 208C being obtained.
S. Ghalem, A. Jutand, Y. Medjour, Eur. J. Org. Chem. 2004, 366–
371; b) M. Ahlquist, G. Fabrizi, S. Cacchi, P.-O. Norrby, Chem.
Commun. 2005, 4196–4198.
[16]a) S. Cacchi, J. Organomet. Chem. 1999, 576, 42–64; b) “Carbopalla-
dation of alkynes followed by trapping with nucleophilic reagents”:
S. Cacchi, G. Fabrizi, in the Handbook of Organopalladium Chemis-
try for Organic Synthesis, Vol. 1 (Ed.: E.-i. Negishi), Wiley-Inter-
science, New York. 2002, pp. 1335–1367; c) C. Amatore, S. Bensa-
lem, S. Ghalem, A. Jutand, J. Organomet. Chem. 2004, 689, 4642–
4646.
[19]In the catalytic reactions, the concentrations of the piperidine and
morpholine were 9.22 and 10.45m respectively.
[20]Y. Takahashi, T. Ito, Y. Ishii, J. Chem. Soc. D 1970, 1065–1066.
[21]D. T. Rosevear, F. G. A. Stone, J. Chem. Soc. A 1968, 164–167.
[17]The reaction of 7b with a mixture of morpholine and alkyne 5 (7b/
morpholine/5=1:2:2) in CDCl3 was not so clear cut. Among the nu-
Received: April 25, 2006
Published online: September 22, 2006
676
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 666 – 676