ative. Using this system (ratio L/Pd: 2) we were pleased to
observe a very efficient transformation at room temperature
even when the catalyst loading was reduced by a factor 5
(
0.2%; Table 1, entry 6). Experimentally, the reduction step
appeared quite exothermic. Changing the solvent from tolu-
ene to THF had a positive effect, reducing the initial exo-
therm, and resulting in a nearly quantitative reaction (90%
by GC, 88% yield of the isolated product) after 2 h at room
temperature (Table 1, entry 7). On the other hand, the other
phosphines did not lead to appreciable catalytic transforma-
tion under similar conditions (Table 1, entry 8).
With an efficient catalyst at room temperature, reaction
conditions were first optimized by varying the ligand/metal
ratio and reaction time, after which the scope of the reaction
was studied (Table 2). In the case of the typically most fa-
Scheme 2. Ligand screening.
[
12]
of chloroarenes at room temperature. This phosphabarre-
lene is a peculiar type of bulky phosphine ligand in that it is
[13]
not electron donating but rather electron accepting, which
is bound to have a profound influence on the energetics of
the elementary steps (OA, TM, and RE) of the catalytic
process. Our goals here are manifold. First, we show that
the cross-coupling between bromoarene and arylzinc deriva-
0
tives can be performed with low loadings of Pd complexes,
which are generated from air-stable [{Pd ACHTUNGTERNN(UGN allyl)Cl} ] and air-
2
[
a]
Table 2. Optimizations and scope of the reaction.
1
2
Entry
Ligand
L/Pd
ratio
t
R
R
Yield GC
ACHTUNGTRENNUNG( isolated)
stable ligand 1 at room temperature. Second, we present
mechanistic stoichiometric investigations which led to the
isolation of a monoligated dimeric complex, an intermediate
in the overall catalytic cycle. Finally, we report computation-
al analysis of the complete catalytic cycle (Scheme 1), which
is in full accord with our experimental results.
In a first set of experiments, the performances of different
commonly used systems were tested in the coupling of an
electron-poor bromoarene (p-Br-PhCN) and an electron-
(h)
1
2
3
4
5
6
7
8
9
1
1
2a
2a
1
1
1
2a
2a
1
2
1
2
2
2
1
1
2
2
2
2
2
1
2
2
2
2
2
24
2
2
16
2
24
2
2
2
p-OMe
p-COOEt
p-CN
85 (74)
60
11
75
90 (88)
30
p-COOEt
32
<5
<5
75 (67)
poor zinc derivative (p-CO Et-PhZnBr). These systems con-
10
p-CH
3
p-OCH
p-CF
3
2
11
12
13
14
1
1
1
1
p-COOEt
p-COOEt
p-COOEt
p-COOEt
p-COOEt
3
85
sist of a catalyst generated in situ from a commercially avail-
o-CN
o-CN
62 (57)
II
able, air-stable Pd complex, Pd
A
H
U
G
R
N
U
G
[b]
2
16
2
48
36
equivalents of a desired phosphine type ligand (Table 1).
Compared with the most efficient “catalytic mixture”
p-OCH
p-OCH
3
3
23
30
2
15
1
1
(
Table 1, entries 1–4), the one generated with phosphabarre-
[a] Reaction conditions: R ArZnBr (5.5 mmol), MeCN (5 mL), R ArBr
lene 1 performed quite poorly (Table 1, entry 5). However,
2
(5 mmol), THF (2 mL), [{Pd ACHNUTGTNERN(GU allyl)Cl} ] (0.1 mol%), THF/MeCN , room
temperature. [b] Transmetalation of the R ArZnBr to R ArZnBr was ob-
served as the major product.
1
2
II
in this configuration, the Pd precursor must be reduced
0
[14]
into a Pd active species before catalysis can begin. This
process appeared to be inefficient with ligand 1. On the
other hand, we have shown previously that the air-stable
complex 3, which is synthesized quantitatively from stoichio-
vorable coupling partners (Table 2, entries 1–4), yields were
slightly higher when using two equivalents of 1 versus Pd
(Table 2. entries 1 and 2). On the other hand, the reaction
was much slower when using an electron-rich phosphine
(Table 2, entry 3), and similar yields were only obtained
after one day (Table 2, entry 4). Most interestingly, the cou-
metric amounts of [{Pd ACHTUNGTERNNNU(G allyl)Cl} ] and ligand 1, could be re-
2
0
duced to an active Pd complex at room temperature with
K CO and PhB(OH) . We thus postulated that the same
2
3
2
complex would be reduced quickly by the organozinc deriv-
1
pling of two electron-poor arene derivatives (R =p-EtOCO
2
and R =p-CN), yielded the best results only when two
[
a]
Table 1. Precursors and solvent screening.
equivalents of ligand 1 were used (Table 2, entry 5). With
one equivalent, the yield reached 30% after two hours, and
only 36% after 16 h, thus suggesting a competitive decom-
position pathway of the catalyst. Note that the use of elec-
tron-rich phosphine 2a did not allow the coupling of these
partners under these conditions (Table 2, entries 8 and 9). In
terms of the scope of the reaction, when the coupling part-
ners possess similar electronic properties, the yields were
good, but decreased drastically when using electron-donat-
ing bromoarene and electron-accepting zinc derivative.
These catalytic results highlighted several important points:
1) the use of two equivalents of phosphabarrelene ligand 1
Entry
Ligand
Palladium
source
Catalytic
loading
Solvent
Yield GC
(isolated)
AHCTUNGTRENNUNG
[%]
1
2
3
4
5
6
7
8
2a
2b
2c
BINAP
1
1
1
2a
Pd
Pd
Pd
Pd
Pd
A
T
N
T
N
U
G
2
2
2
2
2
1
1
1
1
toluene
toluene
toluene
toluene
toluene
toluene
THF
56
61
89
85
40
80
90 (88)
<5
A
H
U
T
N
U
G
A
H
U
G
R
N
U
G
[
b]
A
H
U
T
E
U
A
H
N
T
E
N
N
1
[{Pd
[{Pd
[{Pd
A
T
N
T
E
U
G
2
2
2
]
]
]
0.2
0.2
0.2
A
H
U
G
E
N
N
A
H
U
G
R
N
U
G
THF
1
1
[
a] Reaction conditions: R ArZnBr (R =p-EtOCO) (5.5 mmol), MeCN
2
2
(
5 mL), R ArBr (R =p-CN) (5 mmol), THF (2 mL), L/Pd=2. [b] One
equivalent used.
14390
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14389 – 14393