Ferrocenyl chelated palladacycles as catalysts
Russ.Chem.Bull., Int.Ed., Vol. 59, No. 7, July, 2010
1401
Table 1. Catalytic reaction of bromobenzene (4) with ethyl acryꢀ
late (5) in the presence of bases*
the presence of K PO was slower than that in the presꢀ
3 4
ence of K CO , which may account for the highest yield
2
3
with K PO as a base (see Table 1, entries 1 and 2).
To confirm the stability of palladacycle 3 under our
experimental conditions, we heated it with a threefold
3
4
Entry
Base
Catalyst
Yield of 6
%)
Unreacted
(
C H Br (%)
6
5
excess of ethyl acrylate and K CO in DMF at 140 °C for
1
2
3
4
5
6
7
8
9
Et N
2
2
2
2
2
3
3
3
3
3
1.3
3.3 (16)**
23.0
4.7
3.5
3.8
0—
3.6
0—
<1.0
2.4
1.2
<1.0
3.2
2
3
3
1
2 h. According to TLC data, the final reaction mixture
K CO
2
3
K PO
contained only the starting compound 3.
3
4
Cs CO
Therefore, dimer 2 is an inefficient catalyst for the
Heck reaction under our experimental conditions. At the
same time, compound 3 is a promising catalyst, which
motivated us to examine its properties in more detail, priꢀ
marily in the presence of K CO since this base provided
2
3
KOAc
Et N
1.7
3
K CO
52.8
24.5
4.8
2
3
K PO
3
4
2
3
Cs CO
2
3
the highest yield of ethyl cinnamate 6 in our preliminary
experiments.
10
KOAc
18.8
1.6
*
The reaction time is 7 h; the reactions were not carried out in
A further series of experiments was intended to study
reactions of various aryl halides with ethyl acrylate (5) in
the presence of K CO (DMF, 140 °C) under argon
an inert atmosphere.
*The yield of compound 6 in an argon atmosphere is given in
*
2
3
parentheses.
(
Scheme 3, Table 2).
and K PO ) used in the Heck reaction. In addition, unlike
3
4
Scheme 3
palladacycle 3, the stability of which under various, inꢀ
6
cluding very drastic, conditions is well studied, dimer 2
was always used in reactions with vinyl ketones under mildꢀ
4
,5
er conditions (Scheme 2).
Scheme 2
Reagents and conditions: 3, K CO , DMF, 140 °C.
2
3
Compound
Ar
X
Br
I
Compound
10,15
11,16
Ar
X
I
Br
Br
4
8
9
,6
,13
,14
C H
4ꢀMeC H4
4ꢀMeOC H4
1ꢀNaphthyl
6
5
5
6
C H
6
6
4ꢀMeC H4 Br
12,17
6
It can be seen in Table 2 that the yields of ethyl cinꢀ
namates are sufficiently high, the conversion of the startꢀ
ing aryl halides being complete in 10 h. The exception is
1
ꢀbromonaphthalene (12) (Table 2, entry 6), which was
the sole reagent that was consumed incompletely over that
period of time. This is probably due to mutual steric hinꢀ
drances of the catalyst and aryl halide.
To sum up, we demonstrated that ferrocenyl palladaꢀ
cycle 3 is an efficient catalyst for the Heck reaction.
For the ferrocenyl palladacycles under study, two alterꢀ
native catalysis pathways can take place: associative,
without cleavage of the carbon—palladium σꢀbond, and
dissociative, when this bond breaks down and the cataꢀ
lytically active species contains no palladium—carbon
bond. The dissociative mechanism for palladacycle 3
can confidently be discarded. The sole possibility for
the metal atom in complex 3 to participate in catalysis is
its outerꢀsphere coordination with aryl halide and/or
the vinyl group of acrylate, which initiates the associaꢀ
tive mechanism with an increase in its coordination numꢀ
ber to 5—6.
Reagents and conditions: PhMe, Et N, 110 °C.
3
To study the stability of dimer 2 and its derivatives in
our experiments, we heated compound 2 with an excess of
ethyl acrylate in DMF in the presence of K CO or K PO
at 140 °C for 7 h. However, the expected product 7 was
not detected even in trace amounts, while the starting
dimer 2 decomposed completely even at 100 °C (TLC,
ethyl acetate as an eluent). Apparently, the Pd atom in
dimer 2 is rapidly replaced by ethyl acrylate in DMF and
the product decomposes on further heating. It should be
noted that visual resinification of the reaction mixture in
2
3
3
4